-
Open Access Article
1 - Deep Learning Algorithms in Super-Resolution Images
Bahar Ghaderi Hamid AzadImage super-resolution is one of the important image processing processes to increase the resolution of images and videos. In recent years, methods based on deep neural networks for super-resolution have seen significant progress. The aim of this paper is to provide a c MoreImage super-resolution is one of the important image processing processes to increase the resolution of images and videos. In recent years, methods based on deep neural networks for super-resolution have seen significant progress. The aim of this paper is to provide a comprehensive review on recent developments in super-resolution image using deep learning approaches. In this article, while introducing the concepts of image super-resolution, the common deep learning algorithms for super-resolution and the applications of super-resolution have been investigated. In addition, the set of databases and evaluation criteria are described. This article can open the way for image processing researchers in the super-resolution process. The authors’ effort has been to explore all aspects of this process. Manuscript profile -
Open Access Article
2 - A new method based on texture analysis for the classification of automatic detection of breast microcalcifications of mammography images
Zahra Maghsoodzadeh Sarvestani Jasem Jamali mhdi taghizadeh Mohammad h FatehiMammography is a diagnostic technology used in screening programs to find breast cancer early. By using two techniques for image enhancement and highlighting breast tissue microcalcifications for the desired areas by regional ROI based on fuzzy system and also Gabor fil MoreMammography is a diagnostic technology used in screening programs to find breast cancer early. By using two techniques for image enhancement and highlighting breast tissue microcalcifications for the desired areas by regional ROI based on fuzzy system and also Gabor filtering method, the study's objective was to assess the viability of automatic separation of images of breast tissue microcalcifications and to assess its accuracy. The decision tree classification algorithm is used to categorize the clusters of breast tissue microcalcifications after the clusters have been identified. The samples that are thought to have microcalcification are next highlighted and masked for segmentation, and in the last step, tissue properties are extracted. Then, it was possible to distinguish between benign and malignant forms of segmented ROI clusters with the aid of an artificial neural network (ANN). The results of this work show a high accuracy of 93% and an improvement of sensitivity of 95%, which shows that the presented solution can be reliably applied to detect breast cancer.. Manuscript profile -
Open Access Article
3 - Brain Tumor Detection in Magnetic Resonance Imaging by Deep Convolutional Neural Network
Mitra Afsarinejad Nabiollah Shiri Ramin BaratiIn this paper, brain tumor detection is addressed through the application of advanced deep-learning techniques. The approach involves the development and training of a comprehensive convolutional neural network (CNN) architecture. Leveraging an extensive dataset of brai MoreIn this paper, brain tumor detection is addressed through the application of advanced deep-learning techniques. The approach involves the development and training of a comprehensive convolutional neural network (CNN) architecture. Leveraging an extensive dataset of brain magnetic resonance imaging (MRI), the proposed model expresses its proficiency in the classification of normal brain tissue and tumor-affected regions. The architecture encompasses multiple layers, including convolutional, batch normalization, and pooling layers, culminating in a robust classification layer. Through rigorous training and optimization, the introduced CNN achieves a high level of accuracy in brain tumor classification. The effectiveness of the proposed model is showcased through comprehensive experimentation, demonstrating its potential to significantly contribute to the medical field’s efforts in precise brain tumor diagnosis. Manuscript profile -
Open Access Article
4 - A method based on deep neural network optimized with Huffman algorithm and meta-heuristic algorithms for medical image compression and reconstruction
Mohammad Hossein Khalifeh Mehdi Taghizadeh Mohammad Mehdi Ghanbarian جاسم جمالیThis research makes use of two different approaches to compress medical images for long-term purposes. In the first method, images are compressed using the Huffman cipher and then simplified using a hierarchical modeling based on a neural network-designed categorization MoreThis research makes use of two different approaches to compress medical images for long-term purposes. In the first method, images are compressed using the Huffman cipher and then simplified using a hierarchical modeling based on a neural network-designed categorization. A prediction strategy based on deep neural network training is employed in the second method. This technique uses a trained neural network to infer the locations of individual pixels, hence reducing the amount of data required to describe a picture. Huffman compression encryption is used on the leftover data. An enhanced spatial filtering technique is used to decode the picture data, and the wild horse optimization and gray wolf optimization meta-heuristic algorithms are then used to produce a rebuilt image. Without compromising compression efficiency, this allows for a more realistic application of the suggested solutions in non-deterministic contexts. The suggested approaches allow for picture simplification, which has resulted in faster decoding. Structural similarity index modulation, time and peak signal-to-noise ratio have been improved by an average of 2, 30.1 and 15.15%, respectively. The suggested algorithms were able to compress medical photos with very high quality level, as compared to the current deep learning-based methods. Manuscript profile -
Open Access Article
5 - Predicting stock prices using data mining methods.
Mojtaba HajigholamiThis article discusses data mining methods for predicting financial markets and analyzing sustainable development in financial matters. It also examines the impact of using data mining methods in the stock market and their effectiveness in this area. The research introd MoreThis article discusses data mining methods for predicting financial markets and analyzing sustainable development in financial matters. It also examines the impact of using data mining methods in the stock market and their effectiveness in this area. The research introduces a machine learning approach that generates information using publicly available data and uses this information for accurate prediction. It also explores various data mining methods relevant to financial market analysis, focusing on predicting stock market movements and trends. The study demonstrates that due to the dynamic and variable nature of financial markets influenced by economic, political, and social factors, the use of machine learning and data mining methods can lead to more accurate predictions of stock price movements. Given the extensive and complex data in financial markets, data mining methods have the potential to discover hidden patterns and determine relationships between various variables. Various machine learning algorithms such as artificial neural networks, support vector machines, and random forests, alongside statistical analyses, help improve the analytical capabilities of analysts and investors in making economic decisions. Furthermore, the use of big data and complex analyses has contributed to the development of intelligent trading strategies that can help optimize returns on investments. For example, analysts can enhance the accuracy of their predictions by incorporating sentiment data from social networks into their models. The study emphasizes that sustainable development in financial markets requires a deeper understanding and more precise analysis of data, ultimately leading to stronger data-driven decision-making and trading processes. Manuscript profile -
Open Access Article
6 - Improved analysis of LUG file-related bulk data using LLG
Babak Nikmard Azin Pishdad Golnaz Aghaee Ghazvini mehrdad abbasiNowdays, organizations generate a significant volume of log files that require processing for condition checking, debugging, and anomaly resolution. Outsourcing such processing is not suitable due to the need for real-time processing and security maintenance. Given the MoreNowdays, organizations generate a significant volume of log files that require processing for condition checking, debugging, and anomaly resolution. Outsourcing such processing is not suitable due to the need for real-time processing and security maintenance. Given the multitude of different software and services, organizations face a substantial volume of production logs that should be processed rather than deleted or ignored. In the traditional approach, experts manually check the logs daily. This, on one hand, slows down the process, increases the time and inaccuracy, and, on the other hand, results in a high hiring cost due to the need for an expert force. This article introduces a solution that employs generative neural networks to establish a local structure for log analysis within the organization. The process involves retrieving and parsing text files from various sectors, segmenting them into manageable portions, embedding them, and storing them in a vector database. In this structure, a trained individual without special expertise can quickly access necessary information using appropriate prompts from a local language model available at any time. As a result, three overarching goals are achieved: maintaining security, increasing the speed of analysis, and reducing human resource costs. Manuscript profile -
Open Access Article
7 - Future Studies for Presenting Model of Personal Differentiation Role on Guerilla Advertising with Using Perceptron Artificial Neural Network
tara taefi vahidreza Mirabi Ghasemali Bazaee soheil sarmad saeidiGuerilla Advertising in Contrast of Definition Pattern, Attempt to Use Simplicity and Flexibility Idea For Penetrate and Attract In Customer Heart to Have Maximum Profit With Minimize their Cost. Personal Differentiation Research is so important for result some side of MoreGuerilla Advertising in Contrast of Definition Pattern, Attempt to Use Simplicity and Flexibility Idea For Penetrate and Attract In Customer Heart to Have Maximum Profit With Minimize their Cost. Personal Differentiation Research is so important for result some side of this Attitude Made by Attention on personal personality. The goal of this research is Future Studies Presenting for Presenting Model of Personal Differentiation Role on Guerilla Advertising. The Purpose is Exploratory, the Method is Descriptive Survey. Two Indicator Recognize such as Attitude with psychology Motivation and technical dimension, personality With Myers-Briggs Indicator Test. Population is Go sport Customer, use classify method to choose the sample, number of Sample with Cochran Formula 277 person. For data collecting have2 Specific and closed Questioner including standard MBTI Questioner with 70 Question and another one with 25 Closed And Specified Questioner Made by Researcher , with use Perceptron Artificial Neural Network by (MATLAB Software). the result of Neural Network Get Fit Model With 92%Coefficience Confidence to Define 4 clearly strategic of Interdependencies between Guerilla Advertising and personality, Attitude Indicator for Predict and analysis the new input and select the suitable strategy. Manuscript profile -
Open Access Article
8 - Forecasting of the Students’ Performance in Military Higher Education Using Artificial Neural Network Prediction Algorithm (Case study: A military organization)
Mohammad Fallah Hamideh ReshadatjooBackground: One of the basic issues in a country's higher education system is the foundations of the quality of graduates’ and university students’ performance, which make up two of the seven major issues in the field of quality in higher education and is im MoreBackground: One of the basic issues in a country's higher education system is the foundations of the quality of graduates’ and university students’ performance, which make up two of the seven major issues in the field of quality in higher education and is important in incorporating multiple components in improving the quality of the higher education system of each country, and any ambiguity in it, especially in military higher education, which has a higher sensitivity, will lead to irreparable consequences. Purpose: The main objective of this paper is to forecast the performance of military higher education students using the artificial neural network prediction algorithm. In addition, the main components of student performance quality have been studied. Method: In this paper, using predictive artificial neural network prediction algorithm, forecasting the quality of students' performance in three phases of learning, validation and neural network test was performed. The statistical society consists of faculty members of Shahid Sattari Air University, students and graduates of this university, as well as the members of the Office of Strategic Studies and Theoretical Research, were then interviewed using a semi-structured interview and a researcher-made questionnaire. Finally, MATLAB software was used to model the neural network. Results: Using artificial neural network algorithm, a model with accuracy of 85.5% was designed and tested. Conclusion: By using artificial neural network algorithm and modeling the quality of students' performance, we can accurately predict the quality of the graduates' performance in the Air Force organization. Manuscript profile -
Open Access Article
9 - Expanding the Application Models Box Jenkins, Artificial Neural Network and Adjusted Exponential forecasting Social Phenomena (Case study: forecasting of marriage and divorce in Ilam)
Mohammadreza Omidi Nabi Omidi Ardashir Shiri R. MohammadipourOne of the most important tools in the hands of managers and experts to make strategic decisions is Methods of forecasting and futures. Despite the development of prediction methods, but less likely to use these methods in predicting social phenomena such as marriage, MoreOne of the most important tools in the hands of managers and experts to make strategic decisions is Methods of forecasting and futures. Despite the development of prediction methods, but less likely to use these methods in predicting social phenomena such as marriage, divorce and population growth are discussed. In this study, using data from marriage and divorce between the years 1992 to 2013 in Ilam province to forecasts, the number of these phenomena using models Box Jenkins, Artificial Neural Network and Adjusted Exponential has been studied for years to come. The results showed that the prediction accuracy Box Jenkins model to predict the number of marriages and Artificial Neural Network model to predict the number of divorces is more than any other prediction methods. The predicted values showed that the proportion of marriages end in divorce in Ilam province between the years 2014 to 2018 following the gentle slope, to reduce the move. Manuscript profile -
Open Access Article
10 - A comparison between the power of artificial neural network models and dynamic neural network in predicting exchange rate: an application of wavelet transformation
Mohammad Ali Khatib Semnani Manijeh Hadinejad Roxana KhoshouieThe present study is an attempt in applying the combination of dynamic neural network and decomposition of wavelet in order to make possible the selection of an optimized pattern for predicting considered variable. For the purpose of research, monthly time series of exc MoreThe present study is an attempt in applying the combination of dynamic neural network and decomposition of wavelet in order to make possible the selection of an optimized pattern for predicting considered variable. For the purpose of research, monthly time series of exchange rate from April 1998 to December 2012 were used including 177 observations from which 150 observations were used for modeling purpose and 27 observations were used for simulation or in other words for presenting predictions out of samples. The findings of present study imply that firstly, dynamic neural network models compared to feed-forward multilayer neural networks have better performance in predicting exchange rate out of sample, based on both criteria for prediction error calculation: MSE & RMSE and secondly, applying wavelet decomposition technique improves prediction results of mentioned models based on both criteria. The third point is that among mentioned models, the best result belongs to predictions obtained from dynamic neural networks based on decomposed data by wavelet technique. Therefore, applying this combination of models as an optimized combination is suggested to monetary researchers, analysts and decision makers of country. Manuscript profile -
Open Access Article
11 - Expanding the Application Models Box Jenkins, Artificial Neural Network and Adjusted Exponential forecasting Social Phenomena (Case study: forecasting of marriage and divorce in Ilam)
Mohammadreza Omidi Nabi Omidi Ardeshir Shiri Rahmatullah MohammadipourOne of the most important tools in the hands of managers and experts to make strategic decisions is Methods of forecasting and futures. Despite the development of prediction methods, but less likely to use these methods in predicting social phenomena such as marriage, d MoreOne of the most important tools in the hands of managers and experts to make strategic decisions is Methods of forecasting and futures. Despite the development of prediction methods, but less likely to use these methods in predicting social phenomena such as marriage, divorce and population growth are discussed. In this study, using data from marriage and divorce between the years 1992 to 2013 in Ilam province to forecasts, the number of these phenomena using models Box Jenkins, Artificial Neural Network and Adjusted Exponential has been studied for years to come. The results showed that the prediction accuracy Box Jenkins model to predict the number of marriages and Artificial Neural Network model to predict the number of divorces is more than any other prediction methods. The predicted values showed that the proportion of marriages end in divorce in Ilam province between the years 2014 to 2018 following the gentle slope, to reduce the move. Manuscript profile -
Open Access Article
12 - Bankruptcy Prediction Using Artifical Neural Networks with Camparsion to the Altman Model
M.R. Setayesh D. Ahadianpoor ParvinThis research has been done under title: Bankruptcy Prediction using Artificsl Neural Networks withcamparsion to the Altman Model.The goal of this study is to provide exact explanation and presentation of theoretical basis of research andmeasurement of usefulness bankru MoreThis research has been done under title: Bankruptcy Prediction using Artificsl Neural Networks withcamparsion to the Altman Model.The goal of this study is to provide exact explanation and presentation of theoretical basis of research andmeasurement of usefulness bankruptcy financial models. We presented the research hypotheses in order toprovide suitable scientific context for the study.Hypothese 1: Artificsl Neural Networks and Altman models are suitable instrumental for prediction ofbankruptcy.Hypothese 2: In prediction of bankruptcy one firm, have significant difference the resultsof this two models.The means of the research statements (Balance sheets, Income statement, cash flow statement) of thecompanies which were accepted in Tehran Stock Exchange. The library method was employed in datagathering. Statistical population of research includes active companies whose financial statements areaccessable in Tehran Stock Exchange. The statistical sample of the research includes active companies inproductive industries, from 1379 to 1384.In order to analysis data, We used statistical metods of nonparametric binomial, and for cointegrationsignificant difference two models employed wilcoxon signed- rank test and sign test for hypothese 2. Afteranalyzing the data the results gained id confirmed and supported by above tests Manuscript profile -
Open Access Article
13 - Modeling and forecasting electricity production and Consumption in Iran
Mohammadreza Omidi Nabi Omidi Heshmatolah Asgari Meysam Jafari EskandariDue to the relatively high growth of energy consumption in the country, the future of research in the field of electrical energy as an important intermediate inputs in industrial production and as a final good And the necessary domestic and commercial sector, the requir MoreDue to the relatively high growth of energy consumption in the country, the future of research in the field of electrical energy as an important intermediate inputs in industrial production and as a final good And the necessary domestic and commercial sector, the requirements of law enforcement agencies in the field of production and consumption of electricity. Review and forecast electricity consumption and production managers a valuable factor in the power industry for strategic decision making. In this study, using time-series production and power consumption between the years 1967-2013 and deployment of predictive models Box Jenkins, artificial neural network and gray system in addition to the forecasts for the coming years using the standard average percentage of errors the accuracy of prediction methods were also studied villages. The results showed that the highest accuracy in the prediction of Box Jenkins methods and artificial neural network to predict the power consumption is the highest accuracy. The predicted values showed a decreasing ratio of production to consumption in Iran is relatively constant desire and The electricity production in Iran in 2019 to 318 843 million kW per hour and power consumption to be 260,645 million kWh, Which can be modified using modern methods of production and consumption patterns towards increased production to consumption. Manuscript profile -
Open Access Article
14 - Provide financial policy by predicting financial statement fraud
Seyed jalal Ahmadi Khosrow Faghani Makrani Naghi FazeliBackground: Management responsibility is creating the right organizational climate in which fraud is the worst crime. methods of identifying fraud play an important role in preventing fraud. Objective: To provide financial policy to management in predicting financial fr MoreBackground: Management responsibility is creating the right organizational climate in which fraud is the worst crime. methods of identifying fraud play an important role in preventing fraud. Objective: To provide financial policy to management in predicting financial fraud by using neural network data mining Research method: Descriptive-applied research method and time domain is also from 2008 to 2017. In this study, financial ratios for both fraudulent and non-fraudulent samples and network data mining were analyzed. Pearson's correlation coefficient was then examined for the model linearity for financial ratios and the elimination of independent correlated variables. In the next step, the neural network method was used to provide financial policy to management regarding the prediction of financial statement fraud. Findings: The decision tree method is effective in providing financial policy to management in predicting financial statement fraud. Conclusion: Since the decision tree method has 65.4% correct forecast, it can be effective in providing financial policy to management to predict fraud. Manuscript profile -
Open Access Article
15 - Estimating Energy Consumption of Educational Spaces Using Artificial Neural Networks (ANNs)
مهسا فلاح نیاSize of classroom’s windows has significant effects on both comfort level of users and electricity consumption for lighting. Moreover, windows are the main source of energy loss in classrooms in both cooling and heating Sectors. Considering the large number of education MoreSize of classroom’s windows has significant effects on both comfort level of users and electricity consumption for lighting. Moreover, windows are the main source of energy loss in classrooms in both cooling and heating Sectors. Considering the large number of educational buildings and long life cycle of such buildings, choosing proper window size is crucial for energy saving in sustainable architectural design. Despite the role that windows have in energy consumption, the literatures are surprisingly limited in providing detailed recommendations for architects in determining the appropriate window size in different climates. Therefore, energy based window design has always been complicated for architects due to the number of involved different components and variables. In order to help the architectural designers, in this paper a new methodology is developed using a well-known artificial intelligence technique. In proposed methodology, a predictive model for energy consumption cost in terms of window to wall ratio (WWR) and the window facing was created using Artificial Neural Network (ANN). The methodology consisted of a limited sets of direct numerical energy simulations for any specific climatic zone to generate the data required for training the ANN. The DOE-2 is suggested in the proposed methodology for direct numerical energy simulations of the daylighting scenarios required for training the ANN. The DOE-2 is a popular and powerful computational model developed with financial support of U.S. department of energy. The trained ANN-based model provides a fast and convenient way of comparing the different daylighting scenarios in designing stage. Indeed, further calculations for direct energy simulations are not necessary and an architect can readily utilize the trained ANN-based model as a powerful tool for forecasting the total energy consumption cost. In order to show the applicability and performance of the proposed approach, 288 daylighting scenarios for a standard classroom in a warm and dry climate, Shiraz-Iran, were simulated to determine the corresponding electric and gas consumption. A square classroom of side 7.4 m is the standard classroom defined by Iranian Organization for Renovating, Developing and Equipping Schools. The DOE-2 is utilized for simulating the defined standard classroom in the study area for estimating the annual gas and electric consumption of the generated scenarios over a 50 years period. The included daylighting scenarios were randomly split into train and test sets. In this study, around 80 percent of data were used for training, and the rest were used to evaluate the performance of the trained ANN. The best training and learning functions for different number of layers and neurons was determined in a trial-error process. Correlation Coefficient (CC), Mean square error (MSE) and Root mean square error (RMSE) are the statistical indices used for training procedure. The best results were obtained with 2 hidden layers and 6 neurons per layer. The 'Levenverg-Marquardt back propagation (trainlm)' and 'perceptron weight and bias learning function (learnp)' were the best training functions found for this research. The results show that the trained ANN can accurately predict the total energy consumption cost (RMSE=0.0811, MSE=0.0066, and CC=0.9672). Manuscript profile -
Open Access Article
16 - Estimating Energy Consumption of Educational Spaces Using Artificial Neural Networks (ANNs)
Mahsa FallahniaSize of classroom’s windows has significant effects on both comfort level of users and electricity consumption forlighting. Moreover, windows are the main source of energy loss in classrooms in both cooling and heating sectors.Considering the large number of educa MoreSize of classroom’s windows has significant effects on both comfort level of users and electricity consumption forlighting. Moreover, windows are the main source of energy loss in classrooms in both cooling and heating sectors.Considering the large number of educationalbuildings and long life cycle of such them, choosing proper window size is crucial for energy saving in sustainablearchitectural design. Despite the role that windows have in energy consumption, the literatures are surprisinglylimited in providing detailed recommendations for architects in determining the appropriate window size in differentclimates. Therefore, energy based window design has always been complicated for architects due to the numberof involved different components and variables. In order to help the architectural designers, in this paper a newmethodology is developed using a well-known artificial intelligence technique. In the proposed methodology, apredictive model for energy consumption cost in terms of window to wall ratio (WWR) and the window facing wascreated using Artificial Neural Network (ANN). The methodology consisted of a limited sets of direct numericalenergy simulations for any specific climatic zone to generate the data required for training the ANN. The DOE-2issuggested in the proposed methodology for direct numerical energy simulations of the daylighting scenarios requiredfor training the ANN. The DOE-2 is a popular and powerful computational model developed with financial supportof U.S. department of energy. The trained ANN-based model provides a fast and convenient way of comparing thedifferent daylighting scenarios in designing stage. Indeed, further calculations for direct energy simulations are notnecessary and an architect can readily utilize the trained ANN-based model as a powerful tool for forecasting thetotal energy consumption cost. In order to show the applicability and performance of the proposed approach, 288daylighting scenarios for a standard classroom in a warm and dry climate, Shiraz-Iran, were simulated to determinethe corresponding electric and gas consumption. A square classroom of side 7.4 m is the standard classroom definedby Iranian Organization for Renovating, Developing and Equipping Schools. The DOE-2 is utilized for simulating thedefined standard classroom in the study area for estimating the annual gas and electric consumption of the generatedscenarios over a 50 years period. Included daylighting scenarios were randomly split into train and test sets. In thisstudy, around 80 percent of data were used for training, and the rest were used to evaluate the performance of thetrained ANN. The best training and learning functions for different number of layers and neurons was determined ina trial-error process. Correlation Coefficient (CC), Mean square error (MSE) and Root mean square error (RMSE)are the statistical indices used for training procedure. The best results were obtained with 2 hidden layers and 6neurons per layer. The 'Levenverg-Marquardt back propagation (trainlm)' and 'perceptron weight and bias learningfunction (learnp)' were the best training functions found for this research. The results show that the trained ANN canaccurately predict the total energy consumption cost (RMSE=0.0811, MSE=0.0066, and CC=0.9672). Manuscript profile -
Open Access Article
17 - A new method for ranking of Z-numbers
M. Matinfar S. ezadiIn this paper we propose a new method for ranking Z- numbers and generalizations. This method is based on the internal structure of the artificial neural network, which suggests that the structure of this network consists of inputs weights and the transfer function line MoreIn this paper we propose a new method for ranking Z- numbers and generalizations. This method is based on the internal structure of the artificial neural network, which suggests that the structure of this network consists of inputs weights and the transfer function linear, nonlinear and sometimes linear and nonlinear. It is shown that the proposed method while possessing the ranking properties for Z -numbers whose components of the limiting part are equal and their confidence interval having the same center of gravity has a more logical ranking than those using the center of gravity. While some of the available methods for Z numbers whose boundaries are equal but not equal to their reliability but have the same focal gravity they rank equally which can not be logical in all cases. Therefore, the proposed method overcomes this problem. In some examples the correctness of the subject is shown. the results are compared with some existing methods. Manuscript profile -
Open Access Article
18 - Presenting a methodology based on the self-organizing maps and multi-layer neural networks for suspected money laundering events at bank branches
Hamid Mahdavi Khokhsarai Mohammadreza Shahriari Fereydon Rahnema Rudpashti Syed Abdullah Sajjadi JagharghGiven the importance of banking systems and the misuse of this platform for money laundering purposes, the urgent need for the implementation of anti-money laundering systems by governments and policy makers in economic affairs is important. Also, due to the growth of t MoreGiven the importance of banking systems and the misuse of this platform for money laundering purposes, the urgent need for the implementation of anti-money laundering systems by governments and policy makers in economic affairs is important. Also, due to the growth of terrorism and organized fraud, and the passage of numerous laws against these cases, the need for these systems is increasing. On the other hand, the complexity of money laundering suspicious behaviors is such that no significant action can be taken to detect money laundering without intelligent and data-driven tools. An important and perhaps practical point in Iran is the proximity of these systems to anti-bribery, fraud, violation and inspection systems, which can be considered as an efficient tool for the bank's inspection unit. This paper presents an approach based on data analysis and processing. In this approach, using self-organizing maps, bank branches are clustered based on similar behaviors, then the process of labeling branches is performed using a linear index. In the next step, using the training of a multi-layer neural network, a model for identifying bank branches in which suspicious money laundering processes take place is introduced. Manuscript profile -
Open Access Article
19 - An efficient one-layer recurrent neural network for solving a class of nonsmooth optimization problems
Mohammad Javad Ebadi Alireza Hosseini Hossein JafariConstrained optimization problems have a wide range of applications in science, economics, and engineering. In this paper, a neural network model is proposed to solve a class of nonsmooth constrained optimization problems with a nonsmooth convex objective function subje MoreConstrained optimization problems have a wide range of applications in science, economics, and engineering. In this paper, a neural network model is proposed to solve a class of nonsmooth constrained optimization problems with a nonsmooth convex objective function subject to nonlinear inequality and affine equality constraints. It is a one-layer non-penalty recurrent neural network based on the differential inclusion. Unlike most of the existing neural network models, there is neither a penalty parameter nor a penalty function in its structure. It has less complexity which leads to the easier implementation of the model for solving optimization problems. The equivalence of optimal solutions set of the main optimization problem and the equilibrium points set of the model is proven. Moreover, the global convergence and the stability of the introduced neural network are shown. Some examples including the L1-norm minimization problem are given and solved by the proposed model to illustrate its performance and effectiveness. Manuscript profile -
Open Access Article
20 - پیش بینی هوشمند نقدینگی دستگاههای خودپرداز برمبنای تقاضای مشتریان
ندا کیانی قاسم توحیدی شبنم رضویان نصرت الله شادنوش مسعود صانعیتزریق بیش از اندازه وجه به خودپردازها موجب تحمیل هزینه اضافی به بانک و کمبود وجه در دستگاهها موجب نارضایتی مشتریان و به خطر افتادن برند بانک خواهد شد. برای این منظور باید در دستگاههای خودپرداز وجه نقد قابل ملاحظهای تزریق شود تا پاسخگوی نیاز مشتریان باشد؛ اما تأکید بر Moreتزریق بیش از اندازه وجه به خودپردازها موجب تحمیل هزینه اضافی به بانک و کمبود وجه در دستگاهها موجب نارضایتی مشتریان و به خطر افتادن برند بانک خواهد شد. برای این منظور باید در دستگاههای خودپرداز وجه نقد قابل ملاحظهای تزریق شود تا پاسخگوی نیاز مشتریان باشد؛ اما تأکید بر این رویه ممکن است سبب رسوب پول در دستگاهها شده و زیانهای اقتصادی برای بانک به همراه داشته باشد. بنابراین، بانکها همانطور که به مدیریت نقدینگی در شعب میپردازند، باید با توجه به شرایط زمانی، مکانی و اقتصادی به مدیریت نقدینگی دستگاههای خودپرداز نیز بپردازند. مهمترین گام در این راستا تشخیص میزان تقاضای وجه نقد مشتریان است. بدین منظور میانگین تراکنشهای 9 ماه سال 95 برای 1377 دستگاه خودپرداز مورد سنجش قرار داده شده و در این مقاله سعی شده است تا با پیدا کردن یک الگوی رفتاری از مشتریان با استفاده از شبکه عصبی سری زمانی پویا (NARX) روند نقدینگی دستگاههای خودپرداز پیش بینی شود. نتایج بدست آمده نشان می دهد که مدل طراحی شده با شبکه عصبی پویا نسبت به مدلهای کلاسیک از کارایی بهتری برخوردار بوده است. Manuscript profile -
Open Access Article
21 - Solving of stochastic Voltaire integral equations by fuzzy artificial neural network method
Hadi Abtahi Hamid Reza Rahimi Maryam moslehVoltaire integral equations as the output of problems in basic sciences and engineering have a special application in advancing the solution of complex problems. One of the most widely used types, which consists of a random process under external motion, is the equation MoreVoltaire integral equations as the output of problems in basic sciences and engineering have a special application in advancing the solution of complex problems. One of the most widely used types, which consists of a random process under external motion, is the equations of random Volta integral. Solving this type of equation has always been a challenge for researchers. On the other hand, with the development of artificial intelligence and the presentation of fuzzy artificial neural network method as a model inspired by the process of thinking and analysis in the human brain, advanced models of algorithms have been designed. Some of these learning algorithms have been used in fuzzy artificial neural networks to solve equations. In this paper, using this method and designing a learning algorithm, the random equations of random Voltaire type is investigated. The method presented in this article, in addition to being more accurate than the previous methods, posseses more speed for solving problem. This topic provides an acceptable level of confidence for researchers when dealing with such issues. Manuscript profile -
Open Access Article
22 - Modeling the quality of water and wastewater treatment using neural networks and hybrid neural networks
Ahmad Jafarian Fatemeh Ghanbary Rahim saneeifardOne of the most important and fundamental factors in the life of living things is water. Therefore, water pollution is a major environmental problem and prevent water pollution and providing smart methods for water treatment is so important. Equipping engineering scienc MoreOne of the most important and fundamental factors in the life of living things is water. Therefore, water pollution is a major environmental problem and prevent water pollution and providing smart methods for water treatment is so important. Equipping engineering sciences with intelligent tools and artificial intelligence in the diagnose quality of wastewater treatments can reduce the errors of the methods. This paper presents a simple and hybrid neural network with statistical logistic regression method for modelling of the output quality of wastewater treatment. The proposed intelligent method plays an important role in the quality of wastewater treatment and can be used by artificial intelligence researchers and environmental engineers. Comparison of the predicted results by simple neural network and hybrid one showed that the efficiency of the hybrid model and it is suitable for our purpose. results of research proved that the new method has the highest efficiency with minimum errors. Manuscript profile -
Open Access Article
23 - A Recurrent Neural Network to Identify Efficient Decision Making Units in Data Envelopment Analysis
A. Ghomashi G. R. Jahanshahloo F. Hosseinzadeh LotfiIn this paper we present a recurrent neural network model to recognize efficient Decision Making Units(DMUs) in Data Envelopment Analysis(DEA). The proposed neural network model is derived from an unconstrained minimization problem. In theoretical aspect, it is shown th MoreIn this paper we present a recurrent neural network model to recognize efficient Decision Making Units(DMUs) in Data Envelopment Analysis(DEA). The proposed neural network model is derived from an unconstrained minimization problem. In theoretical aspect, it is shown that the proposed neural network is stable in the sense of lyapunov and globally convergent. The proposed model has a single-layer structure. Simulation shows that the proposed model is effective to identify efficient DMUs in DEA. Manuscript profile -
Open Access Article
24 - Using Neural Network to Determine Input Excesses, Output Shortfalls and Efficiency of Dmus in Russell Mode
D. Modhej M. Sanei N. ShojaData Envelopment Analysis (DEA) has two fundamental approaches for assessing theefficiency with different characteristics; radial and non-radial models. This paper isconcerned the non-radial model of Russell which is a non linear model. Conventional DEAfor a large datas MoreData Envelopment Analysis (DEA) has two fundamental approaches for assessing theefficiency with different characteristics; radial and non-radial models. This paper isconcerned the non-radial model of Russell which is a non linear model. Conventional DEAfor a large dataset with many inputs/outputs would require huge computer resources in termsof memory and CPU time. Artificial Neural Network (ANN) is one of the most populartechniques for non linear models and for measuring the relative efficiency of a large datasetwith many inputs/ outputs. Also in the last decade researches focused on efficiencyevaluation via DEA as well as using ANN. In this paper we will estimate the input excessesand the output shortfalls in addition to efficiency of Decision Making Units (DMUs) inRussell model through ANN. The proposed integrated approach is applied to an actualIranian bank set; the result indicates that it yields a satisfactory solution.works. Manuscript profile -
Open Access Article
25 - A conjugate gradient based method for Decision Neural Network training
M. Nadershahi A. D. Safi Samghabadi R. Tavakkoli-MoghaddamDecision Neural Network is a new approach for solving multi-objective decision-making problems based on artificial neural networks. Using inaccurate evaluation data, network training has improved and the number of educational data sets has decreased. The available trai MoreDecision Neural Network is a new approach for solving multi-objective decision-making problems based on artificial neural networks. Using inaccurate evaluation data, network training has improved and the number of educational data sets has decreased. The available training method is based on the gradient decent method (BP). One of its limitations is related to its convergence speed. Therefore, decision makers can simply guess the necessary data. In this paper, for increasing the Decision Neural Network training efficiency, a conjugate gradient based method has developed for network training. The key point in decision neural network training is to keep the same structures and parameters of the two sub network (multilayer perceptron) through training process. The efficiency of the proposed method is evaluated by estimating linear and nonlinear utility function of multi-objective decision problems. The results of the proposed method are compared with previous existing method and show that in the proposed method, convergence is faster than previous methods and the results are more favorable. Manuscript profile -
Open Access Article
26 - Designing and presenting a model to determine the effect of macroeconomic and banking variables on the occurrence of asset freezing in the country's banking system
Fateme Davoudi Farkoosh mohammad ebrahim Mohammadpoor zarandi mehrzad minoueiIn this article, the goal is to design and present a model to determine the effect of macroeconomic and banking variables on the occurrence of asset freezing in the country's banking system using meta-heuristic models. The current research is applied in terms of purpose MoreIn this article, the goal is to design and present a model to determine the effect of macroeconomic and banking variables on the occurrence of asset freezing in the country's banking system using meta-heuristic models. The current research is applied in terms of purpose, in terms of research method, correlation analysis type and in terms of overall research design, post-event and retrospective. In order to answer the research questions, the annual data of macroeconomic and banking variables, during the period of 1399-1390, were collected and using the test of regression models in EViews, Smart PLS software and also the neural network model. It was estimated in SPSS Modeler software. The estimation results of the regression model of the first hypothesis in EViews software showed that the economic variables of GDP, unemployment rate and interest rate, consumer price index, currency strength at the error level of one percent and the economic growth rate variable at the error level of ten percent have a significant relationship. They have a dependent variable (asset freezing). Also, the estimation results of the structural model of the first hypothesis in the PLS software are significantly aligned with the output of the Eviuse software. So; The first research hypothesis is confirmed. Also, the results of the regression model estimation of the second hypothesis in EViews software showed that the intra-bank variable of the bank size ratio, return on equity, and the amount of liquidity at the error level of ten percent, and the variables of capital adequacy, return on assets, bank capital, at the error level of one percent. The percentage has a significant relationship with the dependent variable (asset freezing). Also, the estimation results of the structural model of the second hypothesis in the PLS software are significantly aligned with the output of the Eviuse software. So; The second research hypothesis is also confirmed. Manuscript profile -
Open Access Article
27 - Using neural network approach to predict company’s profitability and comparison with decision tree c5 and support vector machine (svm)
Malihe Habibzade Mostafa EzadpourProfit as one of the most important indicators of measuring the performance of the economic unit is one of the important accounting issues that has a high status due to the competitive environment and the importance of quick and proper decision making by managers. There MoreProfit as one of the most important indicators of measuring the performance of the economic unit is one of the important accounting issues that has a high status due to the competitive environment and the importance of quick and proper decision making by managers. Therefore, it is important to analyze the index, factors affecting it and predict profitability. In this regard, the present study was conducted by selecting a sample of 124 observations for the period from 1387 to 1395, based on the basic information of the companies financial statements; the effect of 34 variables on the accuracy of predicting the profitability of the accepted companies by Tehran stock exchange, has been investigated. Tree C5 method was used to determine the significant variables in predicting profitability due to the high ease of understanding of the model. Finally, after determining the effective variables and identifying 8 variables, the accuracy of the predictions was measured using the neural network technique, the C5 decision tree and the backup vector machine (SVM), and the results from these three algorithms were compared. The results of the comparison show that using the c5 decision tree and the 8 variables have the best prediction with accuracy of 93.54%, and then the neural network model is 81.45% more accurate than the supported vector machine (69.35%) and has an error. Manuscript profile -
Open Access Article
28 - Developing a Stock Technical Trading System Integrating MLP Neural Network with Evolutionary Algorithms
Alireza Saranj Ahmadreza Ghasemi Asghar Eram Reza TehraniStock trading systems development using evolutionary algorithms over the past few years has become a hot topic in financial fields. In this paper, an intelligent technical trading system was proposed using a combination of MLP neural network and evolutionary algorithms MoreStock trading systems development using evolutionary algorithms over the past few years has become a hot topic in financial fields. In this paper, an intelligent technical trading system was proposed using a combination of MLP neural network and evolutionary algorithms (i.e., GA, ACOR, and PSO). In order to select the final variables as the selected features, a return comparison of each indicator ratings was used based on tradings. Finally, the performance of each model is tested in comparison with the buy and hold strategy. The results show that the evolutionary learning algorithms significantly outperform the benchmark models in terms of the average return and the hybrid MLP_PSO model outperforms others. Manuscript profile -
Open Access Article
29 - Using Data Mining Approaches to Predict and Answer the Needs of Venture Capital
Farshid Ghasedi Ghazvini Farshad Faezi Razi Farzaneh HeydarpourNowadays the economy of developed countries work based on small and medium enterprises and knowledge-base industries. Researchers experimental findings indicate financing of small and medium enterprises and start-ups with innovative activities have heterogeneity and spe MoreNowadays the economy of developed countries work based on small and medium enterprises and knowledge-base industries. Researchers experimental findings indicate financing of small and medium enterprises and start-ups with innovative activities have heterogeneity and special characteristics to start a business. This heterogeneity and special characteristics of start-ups confront essential problems in financing for them.Hence, for solving this problem, usage and continuity of venture capital process is necessary in order to encouragement and financing of innovative activities. Beside, in this process, venture capital firms in confrontation by pillar of financial markets need to conformity with the market regulations and policies. On the other hand venture capital firms in face of entrepreneurs and innovators confront challenges by how and state of ventures selection based on recognition and assessment of their risks for success or fail prediction of investments. The purpose of this research is response to this investorschallenges that lead investors to make superior evaluation and decision making in their start-ups investments through identification of the effective criteria on venture capital investments and their risk assessment for making trade-offs between them through multi criteria decision making method by usage of data mining and artificial intelligence Manuscript profile -
Open Access Article
30 - Reviews of Manipulating Prices using QDF & ANN-GA Models in Tehran's Stock Exchange
M. Hossein Poustfroush Alireza Naser Sadrabadi Mahmood MoeinaddinIn this study, Quadratic Discriminant Analysis (QDF) model and the hybrid model of Genetic Algorithm based on Artificial Neural Network)ANN-GA) are used to estimate manipulation of stock prices in Tehran Stock Exchange. In this study, first by using screening data metho MoreIn this study, Quadratic Discriminant Analysis (QDF) model and the hybrid model of Genetic Algorithm based on Artificial Neural Network)ANN-GA) are used to estimate manipulation of stock prices in Tehran Stock Exchange. In this study, first by using screening data method, a sample of 345 companies listed in Tehran Stock Exchange were selected and then information about the 'TEDPIX' index, closing price, volatility of closing price and trading volume in the timeframe years 1387 to 1391 were collected. Afterwards the selected companies categorized into manipulated and non-manipulated groups by using duration dependence test and run test. Then with scrutiny of the trend of Tedpix's chart and volume chart of the manipulated group, Start of price manipulation is determined. In next step by using Linear Discriminant Function)LDF) ,Quadratic Discriminant Function)QDF) and Genetic Algorithm based Artificial Neural Network and by using closing price, volatility of closing price and trading volume variables and also using information in range one year before starting manipulation group and in range four years for non-manipulation group, designed models for forecasting manipulation. At the end, the prediction ability of the models was examined. According to the results, the prediction ability of QDF model compared to the ANN model is better. Manuscript profile -
Open Access Article
31 - The Proposed Model For Prediction Of GDP Using With ARIMA, Neural Networks And Wavelet Transform
bita Shaygani Amir behdad Salami Ramin KhochianiForecasting GDP, is one of the most important economic issues and due to its practical applications has attracted a lot of attentions. Methods of time series analysis and nonlinear methods such as neural network models as long as are used to forecast such variables . GD MoreForecasting GDP, is one of the most important economic issues and due to its practical applications has attracted a lot of attentions. Methods of time series analysis and nonlinear methods such as neural network models as long as are used to forecast such variables . GDP's time series is variable that after the decomposition, with wavelet - a powerful tool for processing data- and analyzing the hidden layers, at some levels, has linear behavior and at other levels, has nonlinear behavior.Therefore, the proposed method would be thus that the time series of quarterly GDP for the period 1367 to 1389 using wavelet techniques are decomposed into different scale components. Next, the approximation level (trend) and cycles with linear behavior have predicted with ARIMA model, and cycles with the nonlinear behavior have predicted with neural network model.The results show that the performance of the proposed method is better than the neural network (NARNET) and ARIMA models. Manuscript profile -
Open Access Article
32 - بررسی و پیش بینی اثرات نرخ بهره واقعی و نرخ ذخیره قانونی بر تورم، بیکاری و تولید حقیقی در ایران
سعید ایرانمنش -
Open Access Article
33 - بررسی اثر تولید سبز بر رفتار نرخ ارز حقیقی در اقتصاد ایران و کشور های گروه شانگهای با رهیافت الگوریتم PSO و ARDL و شبکه عصبی پرسپترون چند لایه
سعید ایرانمنش سید عبدالمجید جلایی -
Open Access Article
34 - پیشبینی روند تغییرات قیمت سهام با بهکارگیری شاخصهای تحلیل تکنیکی و استفاده از روش ترکیبی الگوریتم ژنتیک و شبکه عصبی مصنوعی: مطالعه موردی سهام ایران خودرو
زینب آذریان سید مهدی همایونی -
Open Access Article
35 - Estimated Demand for Medical services Case Study of Shiraz and Arsanjan Cities "Generalized Logit and ANN Method
morteza hasanshahiIntroduction: The doctor is a health care provider which is depreciated due to, work, accidents, environmental pollution and age and increases the demand for health services; many factors have an impact on the demand for health services; this study attempts to measure t MoreIntroduction: The doctor is a health care provider which is depreciated due to, work, accidents, environmental pollution and age and increases the demand for health services; many factors have an impact on the demand for health services; this study attempts to measure the impact of 18 of the most important ones. Methodology: The model used is the Generalized Logit and ANN. The population includes those who referred to Shiraz and Arsanjan hospitals in 1994. The sample consists of 100 patients and 100 patient patients (non-patient) and data Collected through questionnaires. Findings: The results of model validity tests including fit of Goodness (Pearson and Deviance index), Parallel Regression, Maximum likelihood, and Newton-Raphson Algorithm indicate that the validity of the model is up to 84% confidence. According to the results, increase of 1% in visit, 2% of demand for services is reduced. The initial health and beliefs has the same interpretation of visit. An increase of one percent in premiums caused a decrease of 3.11% and increases one unit in health index, decrease of 1.3% and a one year increase in age, increase of 10%, and daily consumption of one cigarette, increases 0.04% in demand Health care. Conclusion: According to the results, health, insurance, education and awareness of body anatomy have the greatest impact and smoking and job have the least impact on the demand for health care. With increasing age, education, insurance coverage, awareness and income, the demand for treatment increases. On the other hand, over time, education, awareness and per capita income are rising, so, the demand for these services will increase in the future. Manuscript profile -
Open Access Article
36 - Providing a Model for Assessing Pharmaceutical Industries Supply Chain Sustainability Using Adaptive Neuro- Fuzzy Inference System (ANFIS)
Mahsa Alahyari Nazanin Pilevari Reza RadfarIntroduction: The health system has one of the most complex supply chains in the world to deal with human health and well-being. The pharmaceutical supply chain is part of this system. For this reason, the purpose of this study is to develop a model to evaluate supply c MoreIntroduction: The health system has one of the most complex supply chains in the world to deal with human health and well-being. The pharmaceutical supply chain is part of this system. For this reason, the purpose of this study is to develop a model to evaluate supply chain sustainability in this industry. Methods: First, by reviewing previous studies, indicators related to supply chain sustainability in economic, social, environmental and governance dimensions were extracted. Then, using a survey method to implement the model designed in ANFIS, a questionnaire was used to evaluate the sustainability factors in the case study. Results: The designed ANFISs with 50 training courses achieved acceptable error rates. In the economic and social dimension, the trend of sustainability changes was initially decreasing and then increasing. In the environmental dimension where negative indicators were considered, the trend is quite decreasing. In the governance dimension, the trend of incremental stability changes is. In the present study, the root mean square error (RMSE) is considered as a criterion for model validation. Conclusion: To improve the condition, sustainability must first be evaluated and measured so that the results can be quantified after the improvement measures. Based on the findings, it can be concluded that the model designed in ANFIS is a good tool for assessing sustainability. Manuscript profile -
Open Access Article
37 - Data Mining as an Intangible Model of Information Therapy and Seeking Behaviors in Immune Deficiency Disease Specialists
Sedigheh Mohammadesmaeil Shiba KianmehrIntroduction: This study analyzed the information therapy behavior of immunologists in the country, based on the Cohennon self-organized neural network model. Method: Applied research has been done by descriptive survey method using neural network technique. The tool is MoreIntroduction: This study analyzed the information therapy behavior of immunologists in the country, based on the Cohennon self-organized neural network model. Method: Applied research has been done by descriptive survey method using neural network technique. The tool is a researcher made-questionnaire that was distributed among 149 people. Using MATLAB software, specialists based on the main components of clustering research, and then by removing each of the main sub-components,, the most effective and least effective option was determined. Results: Analysis showed in information retrieval skills; 63.75% of the population are in the first cluster with an average of 29.88 and in the second cluster 36.24% with an average score of 30.22, and the most important component is the use of keywords and terms related to the required information. About ways to get information; 22.14% of the population with an average score of 54.36 in the first cluster, 18.12% of individuals with an average of 48.11 in the second cluster, 14.09% with an average of 43.28 in the third cluster, 16.1% with an average of 0.04 49 were in the fourth cluster and 29.53% of the people with an average score of 53.72 were in the fifth cluster, and the most important way to find information was to use electronic information sources. Based on the use of various information services, 46% of people with an average score of 54.85 in the first cluster, 20.66% with an average of 49.38 in the second cluster and 32.66% with an average of 43.08 in the third cluster and the most important component of information therapy services has been familiarity with various sources and information services in the specialized field. Conclusion: Neural clustering of information therapy behaviors of the study population and the resulting information transactions, in addition to resulting in awareness of the needs and information resources required by users, as an accessible and low-cost method that improves the quality of information of immunodeficiency specialists leads to the provision of more effective medical services to patients, provides the necessary basis for anticipating information-oriented arrangements and decisions to meet the needs and information carriers requested by users of medical databases and provides managers and staff This field, and as an effective strategy with the highest level of possible standards, leads to the discovery of the intangible pattern of information seeking behaviors of health users, and teaches the audience to use information media intelligently. Manuscript profile -
Open Access Article
38 - Segmentation and Profiling of Customers of Arvand License Plate Registered Vehicles Using Neural Network Algorithm of Self-Organizing Maps Model
nazanin abbasi maryam darvishiThe emergence of car import market in Arvand Free Zone in Khuzestan province, has covered a significant share of car purchases in this vast province. The present study was conducted using artificial network algorithm of self-organizing maps model aiming at segmentation MoreThe emergence of car import market in Arvand Free Zone in Khuzestan province, has covered a significant share of car purchases in this vast province. The present study was conducted using artificial network algorithm of self-organizing maps model aiming at segmentation and profiling of customers of Arvand license plate registered vehicles. In order to identify the profile of customers and adopt an appropriate marketing strategy, the statistical population of this study was formed over 70,000 owners of Arvand license plate registered vehicles across the Khuzestan province. Using Morgan table, 384 people were selected as a sample by random sampling method, and a questionnaire was utilized to collect information. Research in the group was carried out with a quantitative method that was applied in terms of purpose, exploratory-survey in terms of nature, and also cross-sectional time wise. Data analysis was performed by self-organizing neural network analysis (SOM). Simultaneously, the demographic, psychological, and behavioral characteristics have also been considered to categorize customers. Based on the results, 3 sections including indifferent, conservative and loyal customers were identified and named. The validity and reliability of the study were also statistically confirmed. The results of this study show that demographic, psychological and behavioral variables have a decisive and special role in customer segmentation of Arvand license plate registered vehicles. Manuscript profile -
Open Access Article
39 - Factors affecting Dividend Payout Ratio and Comparing Forecast Accuracy of Dividend Payout Ratio using Regression Model and Neural Network in Iran Over-The-Counter (OTC) Market
Hamid Reza Chegini Mohsen Hamidian Negar Khosravi PourThis article aims to study the factors affecting the dividend payout ratio and compare the forecast accuracy of neural network and regression models using the data for the companies listed on Iran OTC market. This article also studies the relationship of last year divid MoreThis article aims to study the factors affecting the dividend payout ratio and compare the forecast accuracy of neural network and regression models using the data for the companies listed on Iran OTC market. This article also studies the relationship of last year dividend payout ratio, fixed assets to total assets ratio, current ratio, assets-to-debt ratio, revenue growth, accounting earnings quality ratio, and cash return on assets as independent variables with dividend payout ratio as dependent variable. For hypothesis testing, second-order multiple egression model was employed with a sample size of 50 companies in OTC market during a 5-year period of which their end of fiscal year was from March 20th, 2011 to September 22nd, 2015. The results showed that last year dividend payout ratio, fixed assets to total assets ratio, current ratio, assets-to-debt ratio, and accounting earnings quality ratio have no significant relationship with the dependent variable. Revenue growth and cash return on assets, however, have a positive, significant relationship with dividend payout ratio. Findings also indicate that forecast error of neural network is smaller than that of regression model. Therefore, neural network gives better forecast Manuscript profile -
Open Access Article
40 - Forecasting the Type of Audit Opinions: A Data Mining Approach
محمدحسین ستایش فهیمه ابراهیمی سیدمجتبی سیف مهدی ساریخانیData mining methods can be used to assist auditors with providing audit opinions.The purpose of this research is to forecast the type of audit opinions using data miningmethods and compare the performance of these methods. Artificial neural networks,support vector machi MoreData mining methods can be used to assist auditors with providing audit opinions.The purpose of this research is to forecast the type of audit opinions using data miningmethods and compare the performance of these methods. Artificial neural networks,support vector machines, nearest neighbors and decision tree methods were used toconduct the research. The sample consists of 842 observations between 2001 and 2010.The observations were divided in two groups: one group for training and the other forassessment of the method. A comparison of the performance of methods indicates thatsupport vector machines approach outperforms the other approaches with a predictiveability of 76%. Also measuring type I and type II error rates of each method shows thatthe performance of support vector machines is higher than the other methods. Manuscript profile -
Open Access Article
41 - Development of a Fraud Detection Model Using an Integrated Approach Based on the Factor Analysis Model and the Artificial Neural Network Method in Firms Listed in Tehran Stock Exchange
Jaber Mohammadmoosaee Babak Jamshidinavid Mehrdad Ghanbari Farshid KheirollahiThe main purpose of this research is to propose a fraud detection model using an integrated approach based on the factor analysis model and the artificial neural network method. The network used to predict corporate financial fraud has 17 neurons (selected financial rat MoreThe main purpose of this research is to propose a fraud detection model using an integrated approach based on the factor analysis model and the artificial neural network method. The network used to predict corporate financial fraud has 17 neurons (selected financial ratios) in the input layer and 1 neuron (corporate fraud status) in the output layer. The conversion function used in the output layer is linear and for the middle layer a non-linear sigmoid function is selected. The neural network used in this research is a feed forward neural network with back propagation algorithm. The statistical population of this study is comprised of the companies listed in Tehran Stock Exchange in the time interval from1392 to 1393. Out of these companies, 140 have been selected as the research sample. The Beneish M-Score model has been used in order to classify the companies with the likelihood of fraudulent and non-fraudulent reporting. According to the Beneish M-Score Model, 78 companies were fraudulent in terms of their reports and 62 were non-fraudulent. For the final selection of the input variables (financial ratios) in the artificial neural network, the confirmatory factor analysis model and the principal component analysis model have been used. The results obtained from the aforementioned models have shown that the reported structure of the neural network model has 7 hidden layer neurons and the momentum learning algorithm has been used for training the network. This algorithm was more precise and functioned better than other reviewed structures. Therefore, it was selected as the final adjustment of the neural network.. The obtained results indicated that the artificial neural network method had a higher performance in this regard; in that the precision of classification of fraudulent and non-fraudulent firms and the overall performance of the artificial neural network method was57and 69%,72,73%,61,62% respectively. Manuscript profile -
Open Access Article
42 - Comparative between cost prediction using statistical methods and neural networks
امیر محمدزاده نسرین مهدی پور آرش محمدزادهPrediction of total cost of water helps the Isfahan municipality to optimize thewater usage in its 14 urban zone. The total cost of water, basically, depends ondifferent parameters. Generally, the analytically prediction of the total cost is verydifficult if not impossi MorePrediction of total cost of water helps the Isfahan municipality to optimize thewater usage in its 14 urban zone. The total cost of water, basically, depends ondifferent parameters. Generally, the analytically prediction of the total cost is verydifficult if not impossible. Thus, applying intelligent systems such as neural networkmodels can be a good alternative. In this paper, using multi-layer perceptron neuralnetwork and error back propagation algorithm, the total cost of municipal water in theIsfahan municipality is calculated based on parameters such as per capita populationand area of each urban zone. In this paper, a model for simulation and prediction ofthe annual total cost of water in Isfahan municipality is developed. The simulationmodel is developed using the regression and the neural network model is built usingdata from 2004 to 2009. Finally, the neural network method is selected as the mainsimulation method for forecasting the total cost of water in the 14 urban zones ofIsfahan. Manuscript profile -
Open Access Article
43 - Providing a neural network model to predict the profits of companies listed on the Tehran Stock Exchange and comparing its accuracy with HDZ and ARIMA models
masoud asadi seyedmozaffar mirbargkar Ebrahim ChiraniProfit forecasting is an important criterion for companies and companies listed on the Tehran Stock Exchange must be very careful in forecasting their profits. This study aims to provide a neural network model to predict the profits of companies listed on the Tehran Sto MoreProfit forecasting is an important criterion for companies and companies listed on the Tehran Stock Exchange must be very careful in forecasting their profits. This study aims to provide a neural network model to predict the profits of companies listed on the Tehran Stock Exchange and compare its accuracy with ARIMA and HDZ models. The research method is an applied research in terms of purpose, an inductive research in terms of logic and a quantitative research in terms of data nature. In order to collect data, the basic financial statements of companies in the period 1398-1393 were used. In this study, neural network method was used to predict corporate profits and two models, ARIMA and HDZ, were evaluated. The results show that the rate of data convergence and regression in the first phase and in the HDZ method equal to 0.79087, in the second phase, in the ARIMA method, it is equal to 0.79184, and in the artificial neural network method, it is equal to 0.79464, which has a higher degree of convergence and regression coefficient. Based on the results, it can be seen that the designed neural network has the ability to predict stock price trends using general and industry indicators, and this, in addition to confirming the neural network's ability to predict financial areas and profitability it also confirms strategy of the price forecast on the Tehran Stock Exchange.‏ ‏‏ Manuscript profile -
Open Access Article
44 - A comparison of different heuristic, mathematical, and intelligent methods in urban landscape aesthetic evaluation (Case study: Gorgan city)
Sepideh Saeidi seyed hamed mirkarimi marjan mohamadzadeh abdoulrasoul salman mahiniIn today's era, human interventions have caused chaos in landscape patterns and degradation in landscape quality. Therefore, identifying landscape aesthetic beauty, and also fundamental planning and valuable areas, and proper planning and design in order to protect and MoreIn today's era, human interventions have caused chaos in landscape patterns and degradation in landscape quality. Therefore, identifying landscape aesthetic beauty, and also fundamental planning and valuable areas, and proper planning and design in order to protect and promote the aesthetic value seem to be necessary and unavoidable. In this research, the aim is to investigate the performance of various experimental methods (multi-criteria evaluation using weighted linear combination), mathematical (logistic regression), and intelligent (neural network)) in estimating the suitability of the aesthetic value of Gorgan city. After theoretical studies and determination of effective criteria, mapping and standardization of the criteria were done and finally, the map of aesthetic-value suitability was prepared based on the methods of weighted linear combination, neural network, and logistic regression. In order to evaluate the performance of different methods and choose the optimal method, ground control points and ROC validation methods were used. The results showed that in the map resulting from the weighted linear combination method, a large part of the data was lost as a result of the linear combination of layers and weighting, and the neural network method with intelligent performance and the ability to combine and analyze non-linearly compared to the weighted linear combination method and also performing back and forth analysis compared to the logistic regression method, better separates the value of the studied area. According to the results of this research, it can be concluded that when there is little knowledge about the studied area and it is not possible to conduct field surveys to record valuable points of view, performing the weighted linear combination method can be a solution, but if it is possible to conduct field surveys to prepare a map of real educational samples as a dependent variable, more accurate results can be obtained with the help of the neural network method and logistic regression, more accurate results can be achieved, and in the meantime, the intelligent neural network method has a higher ability to distinguish the values of the environment image. Manuscript profile -
Open Access Article
45 - Comparison of artificial neural network and multivariate linear regression (MLR) models to predict cover percentage Artemisia aucheri from some soil properties
Mansoreh Kargar Zeynab JafarianSoil ecosystems management for different purposes requires accurate and quantitative understanding of the soil characteristics and their processes. This study was aimed to predict Artemisia aucheri cover though some soil physical and chemical properties in Vavsar rangel MoreSoil ecosystems management for different purposes requires accurate and quantitative understanding of the soil characteristics and their processes. This study was aimed to predict Artemisia aucheri cover though some soil physical and chemical properties in Vavsar rangeland, Kiasar, Mazandaran province. Random systematic sampling was used. Five transects with 100 m length and 10 plots 4 m2 on each transect were established. Then cover (%) of A. aucheri and 50 soil sample from 0-15 cm depth was estimated in each plot. Soil properties including soil organic carbon, total nitrogen, EC, water percentage, CaCo3 percentage, soil texture, and pH were measured. Data were divided in two series: a series for analysis including 70% of the data for and 30% for evaluation of customized models. Result showed that soil water, silt and sand percentages were the most important soil properties for prediction A. aucheri cover in the study area. Prediction of the statistical models in the study area resulted in mean error and root mean square error values of 0.25, 0.06 for ANN equation and 0.43, 0.12 for MLR, respectively. Therefore, the ANN model could provide superior predictive performance when was compared with MLR model. Manuscript profile -
Open Access Article
46 - Optimizing the exploitation of the Shahryar plain aquifer by simulating the groundwater flow using the combined modeling method
Nima Salehi Shafa Hossein Babazadeh Fayaz Aghayari Ali Saremi MohammadReza Ghafouri Masoud Safavi Ali PanahdarIn this study, a multi-objective simulation model was presented and six scenarios were defined based on the groundwater budget for optimal exploitation of Shahryar plain aquifer. For this purpose, first, using the GIS based models and using fuzzy and weight methods, the MoreIn this study, a multi-objective simulation model was presented and six scenarios were defined based on the groundwater budget for optimal exploitation of Shahryar plain aquifer. For this purpose, first, using the GIS based models and using fuzzy and weight methods, the regions with high water and low water in the study area were identified in order to prioritize resources and uses. Then the desired scenarios were simulated and evaluated by the (GMS) model. Finally, in order to increase the accuracy of the research results, the (TDS) concentration and the groundwater budget were simulated using the artificial neural network model (ANN). According to the results of fuzzy and weighted methods, the areas of Rabat Karim, then Islamshahr and finally Shahryar were identified as low water areas. Based on the obtained results, the unsteady state budget and verification were calculated as -344.68 and -109.98 (MCM), respectively. The third scenario with a budget of 203.33 (MCM) was chosen as the best scenario and the budget resulting from that, ratio to the budget resulting from the model (GMS) and the neural network for the year 95, has increased by 284.87 percent and 284.83 percent respectively. Also, the concentration of total dissolved solids in the groundwater obtained from the desired models in the entire study period was estimated by 655 and 651 mg/liter on average. The criteria of correlation coefficient and determination coefficient obtained from neural network models for groundwater budget and total data were estimated to be equal to one and for the total dissolved solids concentration of groundwater were estimated to be 0.997 and 0.994, respectively. In the present research, the multi-objective simulation pattern as a comprehensive and practical method by providing new simulation methods has the ability to support several effective scenarios and leads to increase the stability of the groundwater system. Manuscript profile -
Open Access Article
47 - Rehabilitation of Aquatic Ecosystems Based on environmental water rights upstream of Water Reservoirs with Inlet Flow Prediction Approach (Case Study: Taleghan Dam Basin)
Zahra Nafariyeh Mahdi Sarai Tabrizi Hossein Babazadeh Hamid Kardan MoghaddamLimited water resources and increased water demand in recent decades have caused irreparable damage to the country's water resources. One of the important components in surface water optimization and management is long-term and short-term river flow forecasts. The aim o MoreLimited water resources and increased water demand in recent decades have caused irreparable damage to the country's water resources. One of the important components in surface water optimization and management is long-term and short-term river flow forecasts. The aim of the present study is to compare the performance of two Bayesian BN network models with probabilistic approach and MLP neural network. Then selecting the best structural model for flow prediction is another goal of the present study. Monthly meteorological data including precipitation, monthly average temperature, evaporation and. Also, the volume of water transferred from five hydrometric stations entering the Taleghan Dam from 2006 to 2018 was introduced as input data to the models. and runoff to the dam was considered as predictable. Then, with the aim of estimating the best Prediction pattern structure, Input data with different layouts were introduced to the models. In the next step, using the hydrological method of Tennant, The environmental discharge was calculated And the probability of these discharges occurring in the registration data and seventeen patterns in the Easyfit software environment was calculated. Then comparing the selected pattern according to the probability of occurrence and the criteria of the index, Nash-Sutcliffe coefficient (NS), root mean square error (RMSE) and mean absolute prediction error (MAPE) was performed. The best model in BN model with 43.3% similarity and index criteria were estimated to be -3.98, 300, 17.3 and 0.06, respectively. MLP model with 80% similarity and index criteria were introduced as -10.3, 8266, 23.9 and 122.3 in the best model, respectively. As a result, both models performed well in runoff estimation, but comparing the environmental probabilities of the two models in the top five patterns, the BN model has an acceptable accuracy . The basin was also found to be at environmental risk. Manuscript profile -
Open Access Article
48 - Provide a method for targets detection in satellite imagery using deep learning with remote sensing and GIS approach
nader biranvand Mehdi Keykhaee rooin mamaniAutomatic detection of features in different areas according to various objectives, including urban management, military objectives, etc., are one of the most up-to-date and important applications of machine learning today. Combining the Global Geographic System (GIS) w MoreAutomatic detection of features in different areas according to various objectives, including urban management, military objectives, etc., are one of the most up-to-date and important applications of machine learning today. Combining the Global Geographic System (GIS) with images taken from satellite sensors and finally using deep learning methods, which is one of the main branches of machine learning, is a great help to the visible subject. Made the effects in the images using remote sensing science.. At the beginning of this research, the various layers in the proposed algorithm have been comprehensively presented and introduced, and then a new method has been presented in the simultaneous combination of CNN and pooling layers in the algorithm used, which finally It led to a significant reduction in network training time using comprehensive training data with high accuracy and at the same time high volume, which in the end, after entering the fully connected layer to extract and identify the desired goals with acceptable accuracy along with cost-effectiveness. Save time. In this research, using network training through training data, ships in satellite images are detected by creating a fully convoluted FCN network. In order to evaluate the performance and accuracy of the algorithm used in finding and detecting ships in satellite images, by applying this detection algorithm on several other satellite images, Precision, Recall and F1-Score evaluation criteria were used. The values were equal to 100%, 97.61% and 98.83%, respectively, which indicates the accuracy and reliability of the algorithm. Manuscript profile -
Open Access Article
49 - Climate Signals applied to the prediction of evaporation in west of Iran
Enayatolah Rahmati Majid Montazeri Amir Gandomkar Mehran Lashani ZandEvaporation is one of the climatic variables that predict significant role in the planning of water. Due to the relatively high rainfall in areas of West Iran, awareness of the evaporation rate of water in these areas is essential for proper management.The factors influ MoreEvaporation is one of the climatic variables that predict significant role in the planning of water. Due to the relatively high rainfall in areas of West Iran, awareness of the evaporation rate of water in these areas is essential for proper management.The factors influencing rate of evaporation, which are climatic signals according to their role in predicting enables evaporation. Evaporation prediction was performed using artificial neural network model based on climatic signals. the data of evaporation at three synoptic stations and the most important climate signals whit at least 20 years of monthly analysis using NeuroSolution software. The results show that the most Important signals affecting the evaporation areas include; Nina3, Nina1, Sw monsoon, Mei and Nina4.Comparison of observed data with a high correlation between the ANN output data shows. So that the correlation of the Kermanshah station is 71%, Hamedan 82% and Sanandaj 80%.The output data of the neural network and climatic signals, can accurately predict the top 97% of the areas evaporation. Manuscript profile -
Open Access Article
50 - Analysis and investigation of landslide hazard zoning using of hierarchical analysis and artificial neural network models in the southern foothills of the Alborz (Tajrish of Tehran)
Mohsen Ranjbar Asal FadakIn this research, by review of previous works and field works, the Nine factors identifiedeffective factors in landslide hazard and used for analysis risk by GIS software. the occurredland slides in the study area were gathered and rectified by GPS. These Nine maps were MoreIn this research, by review of previous works and field works, the Nine factors identifiedeffective factors in landslide hazard and used for analysis risk by GIS software. the occurredland slides in the study area were gathered and rectified by GPS. These Nine maps werecrossed with the occurred landslide map and Landslides amounts and their areas werecomputed in each class. After determining the rate of each factor (element), land slidezonation was performed in GIS by artificial neural network and AHP Models. The efficiencyof output results of models was assessed by DR and QS indices. The results of DR indexshowed The map was produced using a neural network than maps produced using the analytichierarchy higher accuracy for the study area. Manuscript profile -
Open Access Article
51 - Analysis of lake water level between the climatic signals
Morteza Beitari Khakedi Ebrahim FatahiAmong the important characteristics of each lake, its water level. Understanding of how waterlevel fluctuations in an effective change-related issues including water storage changes - byshoreline construction and environmental issues. In this study, the effect of fluctu MoreAmong the important characteristics of each lake, its water level. Understanding of how waterlevel fluctuations in an effective change-related issues including water storage changes - byshoreline construction and environmental issues. In this study, the effect of fluctuating lakewater discharge and meteorological signals lake basin are discussed. The data used in thisstudy due to high volume in the interval from 1986 to 2008 and from 1951 to 2011 at differentstations. In this study, monthly data signals NAO - Nino1 +2 - Nino3 Nino4 Nino3.4 NOI-NP- PDO - SOI is used. All the above data were obtained from the NCEP. After explaining therelation and the prediction model using artificial neural networks for the same time interval,three months and six months respectively. The results of this model were evaluated andanalyzed. Model output of the neural network software Neurosolutions6 that all stations in thestate, ranging from the time delay of the quarter and half-year delay of the effective signalswing and lake water flows into the lake basin Order 1-NINO3 2 - NINO 3 +4 3 - NINO1 +24-NINO4 is the least of the order of 1 - NAO 2 - NOI 3 - PDO 4 - SOI 5 - NP is. Manuscript profile -
Open Access Article
52 - Development and combination of soft computing and geostatistical models in estimation of spatial distribution of groundwater level
سامان معروف پور احمد فاخری فرد جلال شیریOne of the most important issues in groundwater resources quantitative management is estimating water table level using observation wells network data. The purpose of this study is to estimate the groundwater level using the combination of the geostatistics and soft com MoreOne of the most important issues in groundwater resources quantitative management is estimating water table level using observation wells network data. The purpose of this study is to estimate the groundwater level using the combination of the geostatistics and soft computing methods. Bam Normashir and Rhmtabad plains (Kerman province) with an area of 1928 km2 was selected as a case study of this work. In this study, Kriging and IDW methods were used along with the data driven ANN, ANFIS and GEP models for predicting the spatial distribution of groundwater level, then, the best model was selected for further sampling in the studied region. Data from 65 wells during the period of 2002 to 2011 were used. RMSE, R2, AARE and MAE statistical indices were used for comparing the applied models. Results showed that for all of the models with two input parameters (including longitude and latitude), ANN and IDW models presented the most accurate results with the lowest RMSE (7.138 and 15.062m, respectively) and AARE (33 and 44%, respectively), and the highest R2 (0.606 and 0.596, respectively) for the point and regional estimation of groundwater table level. Finally, ANN-IDW hybrid model was selected for estimation and zoning the groundwater level for the future investigations. Manuscript profile -
Open Access Article
53 - Autoregressive simulation of Zarrinehrud river basin runoff using Procrustes analysis method and artificial neural network and support vector machine models
بهروز سبحانی Mohammad Isazadeh منیر شیرزادRivers flow prediction in river basins has an important role in the operation and correct management of water resources. Determining type and number of estimator models inputs is one of the important steps in rivers flow prediction. Therefore, The Procrustes analysis (P MoreRivers flow prediction in river basins has an important role in the operation and correct management of water resources. Determining type and number of estimator models inputs is one of the important steps in rivers flow prediction. Therefore, The Procrustes analysis (PA) method for determining the number of effective inputs was used. In this study, flow prediction was done using the flow data obtained from the Safakhaneh and Santeh hydrometric stations. The Artificial Neural Network (ANN) and The Support Vector Machine (SVM) models was used for flow prediction. The best estimation of flow is done using the MLP and SVM models in Safakhaneh hydrometric station with RMSE equal to 5.68 (m3/s) and 4.85 (m3/s), respectively, and CC equal to 0.73 and 0.78, respectively. While in Santeh hydrometric station RMSE was equal to 6.44 (m3/s) and 6.36 (m3/s) respectively, and CC was equal to 0.78 and 0.79 respectively for MLP and SVM models. PA-SVM model showed better results than SVM model in estimating Safakhaneh hydrometric stations flow with RMSE equal to 5.45 (m3/s) and CC equal to 0.73 during the test period. The results also indicated that SVM and PA-SVM models estimated the flow of Santeh station with RMSE equal to 6.85 (m3/s) and 7.03 (m3/s) respectively. Basically, results indicated that the Procrustes analysis method can be used as one of the Efficient and suitable methods for determining the number of effective inputs. Comparison of the ANN and SVM results indicated that ANN model has more accuracy than SVM model. Manuscript profile -
Open Access Article
54 - Prediction of soil texture using artificial neural networks
علی محمدی ترکاشوند Elnaz Khanbabakhani Mohammad Ali MahmoodiSoil texture is one of the most important soil properties that affect many physico-chemical properties such as water storage, cation exchange capacity (CEC), soil fertility and soil ventilation. Today, artificial intelligence technology such as neural and neuro-fuzzy ne MoreSoil texture is one of the most important soil properties that affect many physico-chemical properties such as water storage, cation exchange capacity (CEC), soil fertility and soil ventilation. Today, artificial intelligence technology such as neural and neuro-fuzzy networks is used to solve problems in modeling systems and processes. For this purpose, 150 soil samples from a depth of 0 - 15 cm of Gavshan Dam watershed in the Kurdistan province were collected. The geographic locations, height and slope percent of every sampling point were recorded. The particle size distribution of samples was measured in the laboratory using hydrometer method. The longitude and latitude, height, slope percent and soil texture particles of training points were introduced to artificial neural networks to estimate soil texture particles by MATLAB software. The accuracy of model was evaluated by scoring, using statistical indicators such as root mean square error (RMSE), the ratio of geometric mean error (GMER) and correlation coefficient (R2). According to the results, the values for estimating sand and clay are approximately the same and for predicting the silt, less than sand and clay, and 37.0, although less error. The accuracy and accuracy of the model show that the neural network does not have any accuracy and accuracy in estimating the percentage of soil texture components and the soil texture mapping. Manuscript profile -
Open Access Article
55 - Modeling Estimation of Suspended Sediment Rate in Pasikhan River Using Decision Tree Artificial Neural Network
سیدسامان Nasiri Ebrahim Amiri محبوبه shadabiAccurate estimation of sediment transport in rivers due to erosion is an important factor for the management of hydrological projects. Artificial neural networks are of great importance for many reasons, such as the ability to detect patterns, the good relationship betw MoreAccurate estimation of sediment transport in rivers due to erosion is an important factor for the management of hydrological projects. Artificial neural networks are of great importance for many reasons, such as the ability to detect patterns, the good relationship between input and output, and the need for less input data to predict suspended sedimentation. Accordingly, the present study attempts to model the estimation of suspended sediment content in the Pasikhan River using the artificial neural network of the M5 decision tree. The amount of sediment in rivers is subject to many parameters of river geometry, hydraulic flow and sediment properties. For this reason, in this study, it has been tried to reduce the number of effective parameters by first dimensioning the effective parameters on sediment transport capacity. The results showed that the initial decision tree, the M5 tree, does not require pruning and is suitable for use. Three parameters of determination coefficient (R2), mean relative error (ME) and mean squared error (RMSE) were used to evaluate the accuracy of the prediction model. The obtained values for these three parameters were 0.851, 1037.64 and 941.32, respectively, indicating the suitability of these three parameters. Comparison of suspended sediment yield from decision tree model with Pasikhan River measurement data showed that the coefficient of determination was 0.8953 which is a very good value. The results showed that this model is effective in predicting suspended sediment content in the Pasikhan River. Manuscript profile -
Open Access Article
56 - Rainfall-Runoff modeling using Deep Learning model (Case Study: Galikesh Watershed)
Razieh Tatar Khalil Ghorbani mehdi Meftah halghi meysam salarijaziArtificial neural networks (ANN) are one of the data mining methods applied by many researchers in different fields of studies such as rainfall runoff modeling. To improve the performance of these networks, deep learning neural networks were developed to increase modeli MoreArtificial neural networks (ANN) are one of the data mining methods applied by many researchers in different fields of studies such as rainfall runoff modeling. To improve the performance of these networks, deep learning neural networks were developed to increase modeling accuracy. This study evaluated deep learning networks to improve the performance of artificial neural networks in Galikesh watershed and to predict discharge for 1, 3, 6 and 12-month time scale based on 1 to 5 month time scale lags made in precipitation and temperature data. Based on 70% and 30% of the data used for training and test respectively the results demonstrated that in all time steps, the deep learning neural network improved the performance of artificial neural network and on average RMSE decreased in both training and test from 0.68 to 0.65 and 0.84 to 0.73 respectively. Moreover, R-square was increased on average from 0.57 to 0.62 and 0.51 to 0.67 respectively in training and test. We can also denote the effect of temperature on the increase of accuracy of rainfall-runoff modeling. Manuscript profile -
Open Access Article
57 - Comparison of Data Mining Models Performance in Rainfall Prediction Using Classification Approach (Case Study: Hamedan Airport Synoptic Weather Station)
Morteza Salehi Sarbijan Hamidreza DezfoulianBackground and Aim: Rainfall is one of the complex natural phenomena and one of the most crucial component of the water cycle, playing a significant role in assessing the climatic characteristics of each region. Understanding the amount and trends of rainfall changes is MoreBackground and Aim: Rainfall is one of the complex natural phenomena and one of the most crucial component of the water cycle, playing a significant role in assessing the climatic characteristics of each region. Understanding the amount and trends of rainfall changes is essential for effective management and more precise planning in agricultural, economic, and social sectors, as well as for studies related to runoff, droughts, groundwater status, and floods. Additionally, rainfall prediction in urban areas has a significant impact on traffic control, sewage flow, and construction activities. Method: The objective of this study is to compare the accuracy of classification models, including Chi-squared Automatic Interaction Detector (CHAID), C5 decision tree, Naive Bayes (NB), Quest tree, and Random Forest, k-Nearest Neighbors (KNN), Support Vector Machine (SVM), and Artificial Neural Network (ANN) in predicting rainfall occurrence using 50 years of data from the synoptic station at Hamedan Airport. In this study, 80% of the data is used for training the models, and 20% for model validation and the results obtained from the model executions are compared using metrics such as confusion matrix, Receiver Operating Characteristic (ROC) curve, and the Area Under the Curve (AUC) index. To create the classification variable for rainfall and non-rainfall data, based on rainfall data, the days of the year are categorized into two classes: days with rainfall (y) and days without rainfall (n). Data preprocessing is performed using Automatic Data Preprocessing (ADP). Then, Principal Component Analysis (PCA) is employed to reduce the dimensions of the variables. Results: In this study, the PCA method reduces the dimensions of the variables to 5. Also, approximately 80% of the available data corresponds to rainless days, while 20% corresponds to rainy days. The research results indicated that the KNN model with an accuracy of 91.9% for training data and the SVM model with 89.13% for test data exhibit the best performance among the data mining models. The AUC index for the KNN model is 0.967 for training data and 0.935 for test data, while for the SVM algorithm, it is 0.967 for training data and 0.935 for test data. According to the ROC curve for Hamedan rainfall data, the KNN model outperforms other models. Considering the sensitivity index in the confusion matrix, the KNN and SVM models perform better in predicting non-rainfall occurrence for training data. In terms of the precipitation occurrence prediction, the RT and KNN models show better results according to the specificity index. Conclusion: The results demonstrated that for the RT, C5, ANN, SVM, BN, KNN, CHAID, QUEST, accuracy metrics was obtained 86.82%, 89.78%, 89.55%, 89.96%, 88.06%, 91.9%, 88.29%, 87.46%, 91.9%, respectively for training data. Moreover, for test data, the accuracy metrics for this model was obtained 83.82%, 87.9%, 88.12%, 89.13%, 87.12%, 89.13%, 87.12%, 88.19%, 86.93%, 86.76%, respectively. The AUC index in the training data for RT, C5, ANN, SVM, BN, KNN, CHAID QUEST models was 0.94%, 0.99%, 0.94%, 0.94%, 0.93%, 0.97%, 0.93%, 0.89%, respectively. In addition, for the test data, this metric was evaluated 0.89%, 0.89%, 0.93%, 0.94%, 0.92%, 0.90%, 0.92%, 0.88% respectively. As observed, considering accuracy metric and AUC index for training data KNN model and for test data SVM model were more sufficient in rainfall prediction. Manuscript profile -
Open Access Article
58 - Modeling of Physicochemical Properties of Eggplant Slices Coated with Basil Seed Gum during Frying Process
F. Salehi M. A. AsadnahalIntroduction: Fried food products are very popular due to their unique characteristics such as color, smell, taste and desirable texture. Controlling frying conditions and using edible hydrocolloid coatings (gums) is one of the best methods to reduce the oil uptake, moi MoreIntroduction: Fried food products are very popular due to their unique characteristics such as color, smell, taste and desirable texture. Controlling frying conditions and using edible hydrocolloid coatings (gums) is one of the best methods to reduce the oil uptake, moisture retention, and improving the appearance properties of fried foods.Materials and Methods: In this study, different concentrations of basil seed gum (0.0, 0.5, 1 and 1.5% w/w) were used to coat the eggplant slices during deep-frying at 150, 175 and 200°C and the relationship between process parameters and the quality of final product were modeled by genetic algorithm-artificial neural network method.Results: The results of this study showed that coating with basil seed gum reduced the oil uptake of the final product. Coating pretreatment maintained the final product moisture and moisture content of the sample coated with 1.5% gum was higher than the other samples (64.05%). This process was modeled by genetic algorithm-artificial neural network method with 2 inputs that included basil seed gum concentration and frying temperature and 5 outputs that included oil percentage, moisture content, and three main color indexes (yellowness(b*), redness (a*), and lightness (L*)). The results of modeling showed that a network with 4 neurons in a hidden layer and using the hyperbolic tangent activation function can predict the physicochemical properties of fried eggplant slices.Conclusion: The coating containing 1.5% of basil seed gum retained moisture content and reduced oil absorption by the fried samples, and this coating is recommended as a suitable edible coating for coating of eggplant slices before the frying process. Sensitivity analysis results showed that the changes in the concentration of basil seed gum had the highest effect on the lightness index and then on the oil content of fried eggplant slices. The change of frying temperature also had the highest effect on the lightness index of fried samples. Manuscript profile -
Open Access Article
59 - Modeling of Soybean Snack Roasting by Infrared Heating Using Artificial Neural Network (ANN)
H. Bagheri M. Kashani Nejadntroduction: Soybean is recognized as a good source of essential nutrients including protein, oil and several bioactive compounds and soybean has the potential to be used as snack and roasted nut, but most significant factor responsible for such limitation is probably c Morentroduction: Soybean is recognized as a good source of essential nutrients including protein, oil and several bioactive compounds and soybean has the potential to be used as snack and roasted nut, but most significant factor responsible for such limitation is probably considered as the characteristic flavor of soybean. Raw soybean has beany, bitter and astringent flavors. Therefore to improve its consumption, the particular flavor of raw soybean must be removed. Roasting might be considered as one of the best methods for this object. Materials and Methods: In this study, the infrared roaster is designed and soybean has been prepared and roasted according to the experimental condition. In this work, an artificial neural network model was developed for modeling of moisture content of soybean snack during infrared roasting. In order to do this, infrared lamp powers of 250, 350 and 450 W, distance between lamp and sample of 4, 7 and 10 cm and roasting time of 30 min were considered as the inputs and the amount of moisture ratio (MR) was estimated as the output. In addition, three different mathematical models were fitted to the experimental data and compared with the ANN model. Results: Based on these results, artificial neural network model for MR with one hidden layer, Sigmoid function as the transfer function, Levenberg-Marquardt method as the learning rule, 4 hidden neurons, 55% for training subset and 25 and 20 percent for each of validation and test subsets respectively had the best over fitting. The determination coefficient (R2) and root mean square error (RMSE) computed for the ANN model were 0.9992 and 0.01099and for the best mathematical model (Two term model) were 0.9776 and 0.02758, respectively. Conclusion: It was concluded that the artificial neural network model satisfied the work better than the mathematical model concerned with soybean snack roasting. Manuscript profile -
Open Access Article
60 - Optimization of Irradiated Kiwi Fruit Properties Using Response Surface Methodology and Prediction with Neural Network and Regression Method
H. Kiani M. H. Rahmati A. Mohammad-RazdariIntroduction: Application of ionizing radiation as a new method in the preservation ofagricultural products has been introduced since 35 years ago. Gamma irradiation has beenintroduced as a new technique to preserve the foods.Materials and Methods: Kiwifruit cultivar (H MoreIntroduction: Application of ionizing radiation as a new method in the preservation ofagricultural products has been introduced since 35 years ago. Gamma irradiation has beenintroduced as a new technique to preserve the foods.Materials and Methods: Kiwifruit cultivar (Hayward) has been used and the samples with 0(Control), 0.5, 1 and 2 kGy cobalt-60 were gamma irradiated then refrigerated at 3±1°C for aduration of 0, 30.60 and 90 days. In this study Response Surface Methodology and inputparameters consisted of irradiation dose and storage time for optimal conditions for storage ofirradiated kiwis were determined. Finally, using regression and neural network methodsoutput prediction and response surface method were compared.Results : The optimal conditions for storage of irradiated kiwis were determined. Theoptimum point for shelf-life, radiation dose, weight, color parameters L*, a* and b*, ascorbicacid content and pH value, were proposed respectively, (two months, 1 kGy, 48.13 and 45.86,11.03 and 12.79, 29.916, 3.2545). By increasing storage time and radiation dose, the weightof stored samples reduced. By increasing irradiation dose and duration of storage a*parameter decreased. By increasing irradiation dose also the b* parameter reduced but byincreased storage time, L* increased. Similarly, by increasing storage time and radiation dose,pH value increased and ascorbic acid content reduced.Conclusion: The neural network and regression analysis have been employed to predict thechanges in color, weight, pH value and ascorbic acid content. Manuscript profile -
Open Access Article
61 - Optimizing energy consumption in the building sector using neural networks and PSO algorithms (Case study: Bandar Abbas city)
fakhri allahyari Azita Behbahaninia Hossein Rahami Maryam Farahani Samira KhadiviBackground and Objective: Energy consumption in buildings accounts for one third of the country's annual energy consumption, so it is important to provide solutions that can reduce energy consumption in this sector.Material and Methodology: Using questionnaires and expe MoreBackground and Objective: Energy consumption in buildings accounts for one third of the country's annual energy consumption, so it is important to provide solutions that can reduce energy consumption in this sector.Material and Methodology: Using questionnaires and experts’ opinions, effective parameters in energy optimization in Construction Engineering Organization of Bandar Abbas were identified. Variables such as wall and ceiling material, area and type of windows, wall and ceiling insulation thickness were selected. Different modes were investigated with Design Builder software. By training two separate neural networks, how the inputs are connected to two important outputs, which is the amount of energy and carbon dioxide, was obtained. And optimization was performed using the PSO algorithm.Findings: In the obtained model, brick wall with insulation thickness of 5cm, beam roof with insulation thickness of 5cm, triple glazing, ratio of north and east windows to wall in the same direction 70%, ratio of south window to south wall between 41 to 43 percent and the ratio of the west window to the west wall is between 65 to 67 percent, in which the amount of energy and carbon dioxide is the minimum.Discussion and Conclusion: If the energy is selected as target function, the results obtained from the PSO are closely consistent with the optimization results for when the target function is the amount of carbon dioxide. These two functions are in line with each other, and optimizing one will lead to optimizing the other. Manuscript profile -
Open Access Article
62 - Prediction of the results of implementation of air pollution control strategies using the Geo-Artificial Neural Network for Tehran metropolis
Mehran Ghoddousi Farideh Atabi Jafar Nouri Alireza GharagozluBackground and Objective: Predicting the results of the implementation of strategic air pollution control policies is the first and most important challenge for Tehran municipality. The main objective of this study was to define a specific method for assessing the resul MoreBackground and Objective: Predicting the results of the implementation of strategic air pollution control policies is the first and most important challenge for Tehran municipality. The main objective of this study was to define a specific method for assessing the result of urban air pollution controlling strategies in Tehran metropolis using a multi-dimensional decision support system. Method: First, the most appropriate air pollution control strategies were selected based on existing conditions and structures in each zone of the city and then weighed according to selected criteria. Based on the spatial monitoring of air pollution formation patterns in the past and present time, as well as the analysis of their effects, the results of implementing air pollution control strategies were simulated using Geo-Artificial Neural Network models. In the next step, variables of time series and uncertainty variables were simulated for predicting the potential future air pollution patterns and finally, the results of the defined control strategies were evaluated based on spatial thematic layers. Findings: Definition of final clusters of air quality control strategies, weighting and ranking of the selected policies based on defined criteria have been the first findings of this research. Also, extraction of time series zoning based on the data collected during a four-year period, as well as simulation of the baseline scenario models and spatial data layers of their output were among the achievements of this study. Finally, the modeling of the predictive variables, design of the air quality control software and the prediction of the results of the the implementation of air pollution control strategies were presented. The results showed that by applying the Geo-Artificial Neural Network models (GANN), the urban managers could effectively predict the results of implementing the air pollution control strategies. Discussion and Conclusion: The results of this study showed that the spatio-temporal analysis supports the process of evaluation and prediction of the effects of pollution and can be used to determine the best pollution control strategies for the zones affected by air pollution. The final results of GANN models indicate that if the selected strategies are implemented based on the scenarios defined, in the "optimistic scenario", air quality in all areas of Tehran is completely stable and remains healthy, while in the "ordinary scenario" will reduce the level of air pollution up to 70 percent in the autumn and winter season if the selected strategies are implemented compared to the lack of implementation of control plans. The final model of the verification process model also confirmed that the pattern of pollution predicted by the model in each of the urban areas had a proper trend and adaptation compared to the pattern of contamination obtained from the actual results of the field data. Manuscript profile -
Open Access Article
63 - Forecasting Municipal Solid Waste Quantity by Intelligent Models and Their Uncertainty Analysis
Maryam Abbasi Malihe Fallah Nezhad Rooholah Noori Maryam MirabiBackground and Objective: The first step in design of municipal waste management systems is complete understanding of waste generation quantity. Forecasting waste generation is one of the most complex engineering problems due to the effect of various and out of control MoreBackground and Objective: The first step in design of municipal waste management systems is complete understanding of waste generation quantity. Forecasting waste generation is one of the most complex engineering problems due to the effect of various and out of control parameters on waste generation. Therefore, it is obvious that it is necessary to develop approaches to a model such complex events. The objective of this study is forecasting waste generation quantity using intelligent models as well as their comparisons and uncertainty analysis.Method: In this study, Mashhad city was selected as a case study and waste generation time series of waste generation in 1380 to 1390 were used for weekly prediction. Intelligent models including artificial neural network, support vector machine, adaptive neuro-fuzzy inference system as well as K-nearest neighbors were used for modelling. After optimizing the models’ parameters, models’ accuracy were compared by statistical indices. Finally, result uncertainty of the models was done by Mont Carlo technique.Findings: Results showed that coefficient of determination (R2) of artificial neural network adaptive neuro-fuzzy inference system, support vector machine, and K-nearest neighbor models were 0.67, 0.69, 0.72 and 0.64 respectively. Uncertainty analysis was also justified the results and demonstrates that support vector machine model had the lowest uncertainty among other models and the lowest sensitivity to input variables.Conclusion: Intelligent models were successfully able to forecast waste quantity and among the studied models, support vector machine was the best predictive model. Moreover, support vector machine produced the results with the lowest uncertainty the other models. Manuscript profile -
Open Access Article
64 - Evaluation and Optimization of Waste Collection and Transportation System in Urmia by Combining the Response Surface and Artificial Neural Network
Saeid JafarzadehGhoushchi Shabnam Hamidi- MoghaddamBackground and Objective: Optimization of urban waste collection and transportation system has the largest part of waste management costs. Therefore, improving this system and reducing its operating costs as a necessity in urban waste management has always been consider MoreBackground and Objective: Optimization of urban waste collection and transportation system has the largest part of waste management costs. Therefore, improving this system and reducing its operating costs as a necessity in urban waste management has always been considered. Method: Due to the high volatility, changes in the size of the waste, climate change and demographic and substructure tissue, the use of artificial neural network system (ANN) is a suitable method for predicting the production waste size, and on the other hand, for The optimization of the management system of these wastes is also used by the surface response method (RSM). Findings The results of this combined method show that the best combination of factors affecting urban waste transport system was proposed by RSM considering the largest loaded pack with about 26 workers, 10 pickups and 6 trucks. This combination is capable of carrying around 34836 tons of cargo at a cost of 596696000 Rials, which represents a high efficiency over actual values. Also, to predict load, the back propagation algorithm (BP) with 9 neurons in the hidden layer was selected as the best model with a predictive power of 99/19% in prediction of weight and 96/62% in cost prediction. Discussion and Conclusion: The results showed that using the combination of two methods of surface response as a statistical method and artificial neural network as a mathematical method, we can find suitable results for evaluation and optimization of waste collection and transportation system. Manuscript profile -
Open Access Article
65 - Evaluation of Suspended Sediment Load by Sediment Rating Curves and Comparing with Artificial Neural Network and Regression Methods (Case study: Babolrud River Mazandaran Province)
Alireza Mardookhpour Hosein jamasbi Omid AlipourBackground and Objective: In this research the object is prediction of suspended sediment load by and artificial neural network (ANN), Sediment Rating Curves (SRC) and regression methodfor BabolrudRiver in Mazandaran province. Method: The inputs conclude discharge and t MoreBackground and Objective: In this research the object is prediction of suspended sediment load by and artificial neural network (ANN), Sediment Rating Curves (SRC) and regression methodfor BabolrudRiver in Mazandaran province. Method: The inputs conclude discharge and the output is sediments concentration in time series. The input and output of river have positive procedure for (1979-2013) and 75% of data utilized for training and 25% for tests. For training the network, data that recognize issue conditions were selected and some data for testing, Findings: The results show the concentration of sediment suspended load derived artificial neural network and is close together and regression coefficient is 92.8%, while regression coefficient is 83% for sediment rating curves and 90% for statistical method respectively. Discussion and Conclusion: In conclusion, artificial neural network (ANN) has more workability and flexibility for prediction of suspended sediment load to sediment rating curves and statistical methods. Manuscript profile -
Open Access Article
66 - Comparison of Procedure of Artificial Neural Networks, Logistic Regression and Similarity Weighted Instance-Based Learning in Modeling and Predicting the Destruction of the Forest (Case Study: Gorgan-Rood Watershed- Golestan Province)
zeynab moradi Ali Reza Mikaeili-TBackground and Objective: The change in forest cover plays a vital role in ecosystem services, atmospheric carbon balance and thus climate change. The goal of this study is comparison of three procedure of Artificial Neural Network, Logistic regression and Similarity we MoreBackground and Objective: The change in forest cover plays a vital role in ecosystem services, atmospheric carbon balance and thus climate change. The goal of this study is comparison of three procedure of Artificial Neural Network, Logistic regression and Similarity weighted Instance-based Learning (SIM Weight) to predict spatial trend of forest cover change. Method: In this study, land use maps for the periods 1984 and 2012 derived from Landsat TM satellite imagery, was used. Transition potential modeling using artificial neural network, Logistic regression and Similarity weighted Instance-based Learning and prediction based on the best model using Markov chain model was performed. In order to assess the accuracy of modeling, statistics of relative performance characteristic (ROC), ratio Hits/False Alarms and figure of merit was used. Findings: The results show the accuracy of artificial neural network with the ROC equal to 0.975, the ratio Hits/False Alarms equal to 63 percent and the figure of merit is equal to 12 percent. Discussion and Conclusions: Artificial Neural Networks in comparison with Logistic Regression and Similarity weighted Instance-based Learning has higher accuracy and less error in modeling and predicting of forest changes. Manuscript profile -
Open Access Article
67 - Proper Orthogonal Decomposition Performance to Determine the Inputs to the Artificial Neural Network for Prediction of Inflow into Alavian Dam
Saber Moazami Roohollah Noori Mohammad Reza Vesali Naseh Abbas AkbarzadehBackground and Objective: Dams play an important role in development of countries by drinking and agricultural water supply, flood control, hydropower energy supply and recreational purposes. Constructing a dam and making an artificial lake has an important effect on su MoreBackground and Objective: Dams play an important role in development of countries by drinking and agricultural water supply, flood control, hydropower energy supply and recreational purposes. Constructing a dam and making an artificial lake has an important effect on surrounding environment, so being able to forecast the inflow to the dam is an important issue for water resource management. Method: In this study artificial neural network (ANN) was applied to forecast the monthly inflow from Soofichai River to Alavian Dam. Regarding the huge amount of input data to ANN model and for optimizing its application, proper orthogonal decomposition (POD) was used in order to determine the best inputs for ANN model . Finally, the application of ANN and POD-ANN models was evaluated by determination coefficient (R2), mean absolute error (MAE) and average of absolute relative error (AARE). Findings: Results of ANN and POD-ANN models indicated that although ANN output is close to the observed values of inflow to the dam, but it has significant errors. POD-ANN model showed better results than ANN model for high values of inflow. In generall, comparing R2, MAE and AARE values of two models revealed that POD-ANN model had better performance in both calibration and verification steps in comparison with ANN model. R2, MAE and AARE in verification step of POD-ANN model were 0.93, 0.79, and 0.54, respectively. Discussion and Conclusion: Preprocessing data contributes to better performance of POD-ANN than ANN model, especially in high values of inflow. Therefore, it can be concluded that applying data preprocessing and reducing inputs to ANN model enhances its performance. Manuscript profile -
Open Access Article
68 - Estimation of Effluent TSS of Ahvaz Wastewater Treatment Plant Using Inelegant Models
Mojtaba Ghaed Rahmati Hadi Moazed Parvaneh TishehzanIntroduction: The limitation of fresh water resources in the world, especially in arid and semi-arid regions such as Iran, has inevitably led to the reuse of urban wastewater. One of the most important indicators of sewage pollution and comparison with different standar MoreIntroduction: The limitation of fresh water resources in the world, especially in arid and semi-arid regions such as Iran, has inevitably led to the reuse of urban wastewater. One of the most important indicators of sewage pollution and comparison with different standards for reuse or discharge to the water resources is TSS. The present study was conducted in 2016 with the aim of estimation of effluent TSS of Ahvaz wastewater treatment plant using inelegant models. Material and methods: Regard to costly and time-consuming measurement tests of TSS, the capability of multivariate linear regression model, Artificial Neural Network (ANN), and Adaptive Neuro-Fuzzy Inference System (ANFIS) was studied to estimate (TSS) in wastewater treatment plant output by MATLAB and SPSS 21 software. Accordingly, various compounds of sewage quality parameters were evaluated during the 8-year statistical period (2008-2015) as input of models in two daily and monthly modes. Results: The results of the regression model indicated that the maximum R2 for training and verification were 0.75 and 0.67 in daily and 0.68 and 0.66 in monthly period, respectively. The root mean square error (RMSE) in this test was 0.033 and 0.025 in the daily period and 0.053 and 0.053 in the monthly period. The maximum R2 in ANN for training and verification were 0.87 and 0.79 in daily and 0.87 and 0.85 in monthly period, respectively. The RMSE in this test was 0.030 and 0.023 in the daily period and 0.034 and 0.031 in the monthly period. Meanwhile, the maximum R2 in ANFIS for training and verification were 0.91 and 0.83 in daily and 0.89 and 0.87 for monthly period, respectively. The RMSE in this test was 0.026 and 0.025 in the daily period and 0.031 and 0.028 in the monthly period. Conclusion: The results confirmed the application of three models is appropriate, but the ANFIS was considered as a more appropriate model. Manuscript profile -
Open Access Article
69 - Turbidity Removal from Water Using Graphene Oxide as Coagulant and Modeling with Artificial Neural Network
nazila rezania Maryam Hasani Zonoozi motahareh SaadatpourBackground and Objective: In recent years, applications of carbon-based nanomaterials have been developed in various fields such as water and wastewater industry. One of these compounds is graphene oxide (GO), which has attracted a lot of attention due to its high speci MoreBackground and Objective: In recent years, applications of carbon-based nanomaterials have been developed in various fields such as water and wastewater industry. One of these compounds is graphene oxide (GO), which has attracted a lot of attention due to its high specific surface two-dimensional structure and various surface groups. In this regard, the main purpose of this study is to investigate the coagulation properties of graphene oxide in removing turbidity from water and modeling the process using artificial neural network (ANN).Material and Methodology: The samples were prepared by using garden soil and tap water and the GO was purchased in the form of suspension. Jar tests were performed to assess the influence of pH, GO dosage, initial turbidity, settling time and other parameters on the turbidity removal efficiency. In order to simulate the process, Perceptron neural network was used.Findings: Under acidic pH conditions and with increasing the GO dosage from 2.5 mg/L to 40 mg/L, the removal efficiency increased considerably. However, the initial turbidity did not show a clear effect on the process performance. Much of the turbidity removal occurred within the first 10 minutes of the settling time and the flocs’ exhibited higher settling rates at higher GO dosages and acidic pH condition. According to the results obtained from the created ANN model, the coefficient of determination (R2) and the correlation coefficient (R) between the observed and predicted values of the test data were 0.9492 and 0.974, respectively, which reveal the model’s high capability in predicting the process results.Discussion and Conclusion: GO showed high capability in turbidity removal from water. The pH and GO dosage were recognized as the process controller parameters. The ANN data mining model showed good performance in predicting process efficiency. Manuscript profile -
Open Access Article
70 - Investigation of the accuracy of multilayer perceptron network and radial base function in estimating river sediment (Case study: Zayandehrud)
Ramtin Sobhkhiz Alireza MardookhpourBackground and Objective: Estimating the amount of sediment by the river is one of the topics that has been considered by many researchers since the past. Reduction of the dam reservoir capacity because of sediments has different effects on different sections and causes MoreBackground and Objective: Estimating the amount of sediment by the river is one of the topics that has been considered by many researchers since the past. Reduction of the dam reservoir capacity because of sediments has different effects on different sections and causes adverse effects on the water rights that were initially agreed upon, which will impose several economic and specific consequences. This study aims to model and estimate the amount of suspended sediment using existing experimental equations and new methods called black box. Material and Methodology: The discharge (volumetric flow rate) related to Zayandehrud River in Eskandari station, one of the hydrological measuring stations, has been used to estimate the amount of sediment. For this purpose, water discharge and sediment rate are used as input and output, respectively. Findings: According to the obtained results, it is concluded that the RBF network has better performance due to less error in the test stage, but the MLP network seems to have a better performance considering other parameters and the error in the TRAIN stage. Discussion and Conclusion: Finally, after modeling by using neural networks, the Einstein relationship, and the sediment measurement curve, it is inferred that neural networks are more accurate to estimate the amount of sediment. Manuscript profile -
Open Access Article
71 - Development and Improvement of Neural network algorithm and forest cover index (FCD) classification methods in GEOEYE high resolution satellite data. (Case study: Ramsar-Safarood Hyrcanian forests)
Amin Mahdavi Saeidi Sasan Babaie Kafaki Asadollah MattajiBackground and Objective: Due to the high spatial resolution of Geoeye data, due to the wider distribution of pixels, the output maps in Neural network algorithm and Forest cover index (FCD) classification methods are more sensitive and with more pixel detail. Consideri MoreBackground and Objective: Due to the high spatial resolution of Geoeye data, due to the wider distribution of pixels, the output maps in Neural network algorithm and Forest cover index (FCD) classification methods are more sensitive and with more pixel detail. Considering the large amount of information in new sensors, the aim of this study is to develop and improve the performance of more complex classification algorithms for the interpretation of modern satellite data. Material and Methodology: FCD model base classification is based on four main indicators: sensitive to shadow, uncovered soil, vegetation conditions and density, and without the need for a training sample. The Neural network algorithm operates with high sensitivity to the original image bands and the bands created and added to the image, as well as training samples. Training samples were determined in the summer of 2016-2017 from series 5 and 6 of 30 Ramsar watersheds. Finding: Using this method, an accuracy of 24.5% was obtained for the FCD method and 26.2% for the Neural network method. Due to the high resolution of the data used, the output map developed in this method is associated with a high density of polygons. Discussion & Conclusion: Due to the range of pixels in the output maps of the two methods, an extended method has been proposed to produce a more accurate map, due to the high spatial resolution of the Geoeye sensor. In this method, by reclassifying within the maximum frequency range of pixels, the demarcation of polygons in much smaller and more accurate dimensions is considerable. Manuscript profile -
Open Access Article
72 - Prediction of Gas Pollutants Concentration by Means of Artificial Neural Network in Tehran Urban Air
Siamak Bodaghpour Amir CharkhestaniIn this study we applied artificial neural network (ANN) to predict the concentration of air pollutant in Tehran urban air. Because of dangers of air pollution in Tehran city witch causes environmental problems and various respiratory and dermatological diseases and tro MoreIn this study we applied artificial neural network (ANN) to predict the concentration of air pollutant in Tehran urban air. Because of dangers of air pollution in Tehran city witch causes environmental problems and various respiratory and dermatological diseases and troubles especially in children and aged people. This research was set in order to schedule and control this problem in Tehran and other great cities. Statistical data for this purpose were picked up from the concentration of pollutant gases recorded by fixed sensors in Bazar station from 2002 till 2007 (NOX gas). Auto regressive model and time series were used to determine neural network inputs. Current time gas concentration in this model depends on gas concentration of all 7 past days. Therefore, neural network input was concentration of the gas in all 7 past days and neural network output which was the prediction of neural network and the concentration of the gas in current time. Then the model of ANN is deigned by using of MATLAB 7 software and data simulating. Eventually, simulated data has plotted versus real data and it depicted that there is a good result compared with simulated data from ANN. The latter shows less error compare with regression model Manuscript profile -
Open Access Article
73 - Application of GMDH and genetic algorithm in fraction in biogas from landfill modeling
Mohammad Javad Zoqi Mohammad Ghamgosar Mohammad Ghamgosar Saeed FallahiBackground and Objective: In this study, The Group Method of Data Handling (GMDH) type neural networks whit genetic algorithm was applied to estimate the methane fraction in landfill gas originating from Lab-scale landfill bioreactors. In this study, to predict the meth MoreBackground and Objective: In this study, The Group Method of Data Handling (GMDH) type neural networks whit genetic algorithm was applied to estimate the methane fraction in landfill gas originating from Lab-scale landfill bioreactors. In this study, to predict the methane fraction in landfill gas as a final product of anaerobic digestion, we used input parameters such as pH, Chemical Oxygen Demand, NH4+-N and waste temperature. Method: To this Purpose, two different systems were applied for neural network’s data obtained. In system I (C1), the leachate generated from a fresh-waste reactor was drained to recirculation tank, and recycled every two days. In System II (C2), the leachate generated from a fresh waste landfill reactor was fed through a well-decomposed refuse landfill reactor, and at the same time, the leachate generated from a well-decomposed refuse landfill reactor recycled to a fresh waste landfill reactor. leachate and landfill gas components were monitored for 132 days. Findings: The study results indicate that GMDH is able to predict the methane fraction in landfill gas. The correlation between the observed and predicted values for the training data is 0.98 and for the testing data, it is 0.99. Discussion and Conclusion: The proposed method can significantly predict the methane fraction in landfill gas originating and, consequently, GMDH can be use to optimize the dimensions of a plant using biogas for energy (i.e. heat and/or electricity) recovery and monitoring system. Manuscript profile -
Open Access Article
74 - Comparison of Bayesian Neural Networks and Artificial Neural Network to Estimate Suspended Sediments in the RiverS (Case Study: Simineh Rood)
Mohammad Ali Ghorbani Reza DehghaniBackground and Purpose: Simulation and evaluation of sediment are important issues in water resources management. Common methods for measuring sediment concentration are generally time consuming and costly and sometimes does not have enough accuracy. Materials an MoreBackground and Purpose: Simulation and evaluation of sediment are important issues in water resources management. Common methods for measuring sediment concentration are generally time consuming and costly and sometimes does not have enough accuracy. Materials and Methods: In this research, we have tried to evaluate sediment amounts, using bayesian neural network for Simineh-Rood, West Azerbaijan, Iran, and compare it with common artificial neural networks. Monthly river discharge, temperature and total dissolved solids for time period (1354-1383) was used as input and sediment discharge for output. Criteria of correlation coefficient, root mean square error and Nash Sutcliff bias coefficient were used to evaluate and compare the performance of models. Results: The results showed that three models smart estimate sediment discharge with acceptable accuracy, but in terms of accuracy, the bayesian neural network model had the highest correlation coefficient (0.832), minimum root mean square error (0.071ton/day) and the Nash Sutcliff (0.692) and the bias (0.0001) and hence was chosen the prior in the verification stage. Discussion and conclusions: Finally, the results showed that the bayesian neural network has great capability in estimating minimum and maximum sediment discharge values. Manuscript profile -
Open Access Article
75 - A Neural Network Model for Prediction of Tri-Halo-Methane Concentration in Drinking Water
Mohammad Javad Zoqi Mohammad Ali JafariIn this study a neural network model is proposed for modeling tri-halo-methane concentration indrinking water. After training, the neural network model predicts tri-halo-methane concentration basedon input data. Parameters such as pH, Temperature, free chlorine residue MoreIn this study a neural network model is proposed for modeling tri-halo-methane concentration indrinking water. After training, the neural network model predicts tri-halo-methane concentration basedon input data. Parameters such as pH, Temperature, free chlorine residue and TOC were used as inputdata. To validate the proposed method, a case study was carried out, based on the data obtained fromGuilan grand treatment plant (Sangar). The Levenberg-Marquardt algorithm was selected as the bestof thirteen back-propagation algorithms. The optimal neuron number for Levenberg-Marquardtalgorithm is 8 neurons. The performance of modeling was determined. The trends of the forecast andmeasured data were in good agreement. Manuscript profile -
Open Access Article
76 - Comparison of Autoregressive Static and Artificial Dynamic Neural Network for the Forecasting of Monthly Inflow of Dez Reservoir
Mohammad Ebrahim Banihabib Mohammad Valipoor S. Mahmood BehbahaniIn this paper, the capability of autoregressive static and artificial dynamic neural networks models was compared for forecasting of monthly inflow of Dez reservoir. In previous researches, static and artificial dynamic neural networks models have not been compared MoreIn this paper, the capability of autoregressive static and artificial dynamic neural networks models was compared for forecasting of monthly inflow of Dez reservoir. In previous researches, static and artificial dynamic neural networks models have not been compared for above-mentioned propose. In addition, using artificial neural network model as an autoregressive model is innovation point of this research. Monthly flow data of Dez station in Dez River in years1955 to 2001 is used in this research. Data of 42 former years and 5 recent years are used for Training and testing data set, respectively. Different structure for the static and artificial dynamic neural network models were evaluated by comparing the root-mean-square error (RMSE) of the models. First, static and artificial dynamic neural network models were selected in training phase using data from October 1955 to September 1997. Then, using the selected structures, the monthly forecasted inflow of reservoir was compared with observed data from October 1997 to September 2001. Also, two types of radial and sigmoid activation function and the number of neurons in the hidden layer were investigated in this study. Results showed that the best model to forecast the reservoir inflow is autoregressive artificial neural network model associated with the sigmoid activation function and 17 neurons in the hidden layers. Artificial dynamic neural network model with sigmoid activation function can forecast reservoir inflow for 5 years better than static artificial neural networks model Manuscript profile -
Open Access Article
77 - Estimation of Aquifer Qualitative Parameters in Guilans Plain Using Gamma Test and Support Vector Machine and Artificial Neural Network Models
Mohammad Isazadeh seyedmostafa Biazar Afshin Ashrafzadeh Rezvan KhanjaniAbstract Background and Objective: Having information about qualitative and quantitative parameters distribution of groundwater supplies is one of most important parameters in integrated groundwater management. Thus, in this study it has been attempted to determine a pr MoreAbstract Background and Objective: Having information about qualitative and quantitative parameters distribution of groundwater supplies is one of most important parameters in integrated groundwater management. Thus, in this study it has been attempted to determine a proper model and input combination for estimation of quality parameters including electrical conductivity (EC), calcium (Ca) and sodium (Na) ions in aquifers of Guilans plain. Method: In this study, the data from 132 observation wells during 2001 to 2013 were used and artificial neural network (ANN) and support vector model (SVM) were applied. In the first approach, estimations were conducted according to five different combinations, including water level, distance from see, total precipitation of six months and coordinates of observation wells. In the second approach, estimations were conducted based on combination of the selected qualitative parameters of gamma test with combinations of the best input in the first part. Findings: Comparison of the results from the first part indicated that SVM model outperformed the ANN mode in the estimation of Ca, Na and EC parameters. Support vector machine error values for estimating Ca, Na and EC variables at the test period were 1.218 (meq/l), 0.867(meq/l), and 175.742 (µmos/cm), while for artificial neural network these values were 1.268 (meq/l), 0.933 (meq/l), and 186/448 (µmos/cm) respectively. The results from this part showed that adding the distance from see input improves the estimation of models in all cases. In the second part, using gamma test for measuring the nine quality parameters, the best combination of quality parameters was determined to estimate the three parameters: Ca, Na and EC. The results from the second part show that both ANN and SVM models have an excellent performance in the estimation of the three qualitative parameters. ANN model error values in estimating Ca, Na and EC variables in validation period were 0.662 (meq/l), 0.305(meq/l), and 47.346 (µmos/cm), while these values were 0.671 (meq/l), 0.356 (meq/l), and 55.412 (µmos/cm) for SVM model respectively. Obviously, the results from ANN model in this section were better than those from SVM model. Discussion and Conclusion:Results showed that both ANN and SVM models have a great ability in predicting qualitative parameters in the aquifers. Also, in less inputs, the results of SVM model are better than those of ANN model and in more inputs it is vice versa. Results of the second section showed that gamma test is fully practical and accurate in determining the effective input combinations. Manuscript profile -
Open Access Article
78 - Prediction of Carbon Monoxide Concentration in Tehran using Artificial Neural Networks
Hamid Reza Jeddi Rahim Ali Abbaspour Mina Khalesian Seyed Kazem AlavipanahBackground and Objective: Nowadays, air pollution is one of the most important problems almost all over the world. There are many strategies to control and reduce air pollution, one of which is prediction of this event and getting ready to deal with the negative effects MoreBackground and Objective: Nowadays, air pollution is one of the most important problems almost all over the world. There are many strategies to control and reduce air pollution, one of which is prediction of this event and getting ready to deal with the negative effects of it. The aim of this study is to provide a multi-layer structure of artificial neural networks (ANN) for predicting of carbon monoxide pollution at subsequent 24 hours in Tehran metropolis. Method: To predict the amount of CO emissions in near future (subsequent 24 hours), wind speed and direction, temperature, relative humidity, and barometric pressure characteristics are used as meteorological data, and concentration of carbon monoxide is considered as a pollution parameter. To eliminate the noise of data, wavelets transform method and determining the threshold with normal distribution are used before training the ANN. Finally, two neural networks as two general models are proposed and used for modelling. Findings: The results show that the correlation coefficient, index of agreement, accuracy of prediction, and root mean square error for model no. 1 with duplicate data are 0.9012, 0.915, 0.848, and 0.1012 and for model no. 2 with duplicate data are 0.9572, 0.978, 0.963, and 0.0385 respectively. Moreover, the results of listed parameters for model no. 1 with new data are 0.9086, 0.89, 0.885, and 0.0825 and for model No. 2 with new data are 0.8678, 0.928, 0.932, and 0.1163 respectively. Conclusion: Results showed that there is a good agreement between predicted and observed values, hence the proposed models have a high potential for air pollution prediction. Manuscript profile -
Open Access Article
79 - Comparison of Geo-Statistical Methods and Artificial Neural Network in Estimating Groundwater Level (Case Study: Nourabad Plain, Lorestan)
Reza Dehghani Atefeh NooraliBackground and Purpose: Geo-hydrology issues of changes in the water table are very important. Therefore research is necessary to estimate the missing data. Materials and Methods: One of the important methods to estimate the groundwater table is interpolated. Recent dec MoreBackground and Purpose: Geo-hydrology issues of changes in the water table are very important. Therefore research is necessary to estimate the missing data. Materials and Methods: One of the important methods to estimate the groundwater table is interpolated. Recent decades due to the spatial correlation between the values of a variable in a well developed area, geo-statistical science concepts and capabilities in the field of statistics to evaluate and predict the spatial variables expanded. In this study, the interpolation of groundwater level of Noorabad plains in the province of Lorestan, using geo-statistical methods, have been studied and the results were compared with conventional smart as artificial neural network. Measures average absolute error, mean bias error, root mean square error and standard deviation, and the methods used to assess the public. Results: The results showed that the spatial variation of groundwater table co-krigings simple circular model had a mean absolute error (0.0001), mean bias error (0.0347), root mean square error (0.0451m) and standard deviation (20.3) priority than other methods were. Discussion and Conclusions: the results showed a high capacity co-krigings interpolation and prediction groundwater level is minimum and maximum values. Manuscript profile -
Open Access Article
80 - Evaluation of Artificial Neural Network and Multiple Nonlinear Regression Modeling for the determination of Dissolved Organic Carbon
Taher Ahmadzadeh Naser Mehrdadi Mojtaba Ardestani Akbar BaghvandAbstract Background and Objective: Monitoring of organic carbon in water resources is a critical quality index in environmental management, water quality monitoring and drinking water projects. In this study, the performance and applicability of artificial neural networ MoreAbstract Background and Objective: Monitoring of organic carbon in water resources is a critical quality index in environmental management, water quality monitoring and drinking water projects. In this study, the performance and applicability of artificial neural network and multiple nonlinear regression modeling were investigated and optimized for the prediction of dissolved organic carbon. Method: Optimization was performed using backward elimination method with the highest probable correlation coefficient and minimum number of input parameters. Findings: Model verification showed a good agreement between the predicted organic carbon and actual observations. Results showed the acceptable performance of neural network model with the mean absolute error percentage of 7.6% and correlation coefficient of 0.91. Discussion and Conclusion: Further investigations unveiled that although the multiple regression model, with mean absolute error percentage of 8.4% and correlation coefficient of 0.89, seems to be less appealing but its fast run-time and better performance in critical conditions makes it a better choice for the prediction of organic carbon in aqueous solotions with high range of qualitative changes. Manuscript profile -
Open Access Article
81 - Assessment of Intelligent models for Estimating the Electrical Conductivity in Groundwater (Case study: Mazandaran plain)
Isa Hazbavi Reza DehghaniAbstract Background and Objective: Groundwater resources along with surface water supply the needs for municipal, industrial and agriculture uses, and their quantity and quality should be investigated. Salinity is one of the most important parameters in assessing the qu MoreAbstract Background and Objective: Groundwater resources along with surface water supply the needs for municipal, industrial and agriculture uses, and their quantity and quality should be investigated. Salinity is one of the most important parameters in assessing the quality of groundwater. Method: In this study, application of artificial neural networks and Bayesian network in predicting the electrical conductivity in 8 observation wells in Mazandaran plain was investigated. For this purpose, hydrogen carbonate, chloride, sulfate, calcium and magnesium were selected as input and output parameters for electrical conductivity at monthly a scale during 2003-2013. The criteria of correlation coefficient, mean absolute error and Nash Sutcliff coefficient were used to evaluate the performance of the model. Findings: The results showed that artificial neural network model has the highest correlation coefficient (0.989), the lowest mean absolute error (0.019 ds/m) and the highest standard of Nash Sutcliffe (0.970) ranked the first priority in the validation phase. Discussion and Conclusion: The results indicate acceptable capability of artificial neural network models to estimate the electrical conductivity of groundwater. Manuscript profile -
Open Access Article
82 - Determining the optimal method for classification and mapping of land use/land cover through comparison of artificial neural network and support vector machine algorithms using satellite data (Case study: International Hamoun wetland)
amir houshang ehsani Mojtaba ShakeryariBackground and Objective: Images classification is one of the important techniques for interpretation of satellite images that is widely used in survey of earth changes. In the meantime, satellite data has been recognized as the best tool for detection and evaluation of MoreBackground and Objective: Images classification is one of the important techniques for interpretation of satellite images that is widely used in survey of earth changes. In the meantime, satellite data has been recognized as the best tool for detection and evaluation of changes due to its update information, low costs and variety of forms. Therefore, land use/land cover map is one of the most important information required by the environmental managers and planners. On the other hand, in recent years, artificial neural network method has been used widely for the classification of satellite data. The aim of this study is to compare three different methods for land cover classification using 2014 OLI image over a 26-year period. Method: In this study, digital data of OLI (2014) sensor was used in order to optimize image classification method. Initially, the image was corrected in terms of geometry and radiometry in the ENVI software. Then IDRISI software was used for image classification using three different methods: fuzzy artmap, multilayer perceptron artificial neural networks and support vector machine. Finally, land cover maps were classified into five categories: water, vegetation, canebrake, barren lands and saline lands. To evaluate accuracy with the help of user accuracy, producer accuracy, overall accuracy, kappa coefficient and error matrix, the created map was compared with the ground reality map created by GPS, Google Earth images and field observations. Discussion and Conclusion: The results of image accuracy evaluation showed that among the applied methods the fuzzy artmap algorithm had the highest accuracy in classification of satellite data with an overall accuracy of 94.68 and kappa coefficient of 0.91 compared to both multilayer perceptron artificial algorithm with an overall accuracy of 92.99 and kappa coefficient of 0.89 and support vector machine with an overall accuracy of 90.93 and kappa coefficient of 0.85. This study showed that classification of fuzzy artmap artificial neural network algorithm has a high capability to create the land cover map with high accuracy. Manuscript profile -
Open Access Article
83 - Exploring Land Cover Changes in Arak Using GIS and Remote Sensing
Mozhgan Ahmadi Nadoushan Alireza Soffianian Sayed Jamaleddin KhajeddinIntroduction: Land cover changes are among the most important alterations of the Earth’s land surface which affect the environment and environmental processes. Material and Methods: In this study, aerial photos from 1956 and 1972, Landsat TM and IRS-P6 LISS-III im MoreIntroduction: Land cover changes are among the most important alterations of the Earth’s land surface which affect the environment and environmental processes. Material and Methods: In this study, aerial photos from 1956 and 1972, Landsat TM and IRS-P6 LISS-III images obtained in 1980 and 2006 as well as the post-classification method were used to detect land cover changes and to evaluate Arak and its periphery during 1956-2006. At first, geometric correction was done to prepare aerial photos and satellite images. For this purpose, topographic maps at scales of 1:50000 and 1:25000 and nearest neighbor method and resampling method were applied. Root mean square error for all aerial photos and satellite images was less than one pixel. Afterwards, all georefrenced photos were mosaicked and land cover maps with 4 classes (urban areas, vegetated areas, barren lands, and rocks) were generated using visual interpretation of aerial photos. Following satellite images geometric correction, topographic correction was applied to images using DEM and Lambert model. In the next step, thye artificial neural networks classification method was implemented after producing false color composite images and image fusion. Results: Land cover maps in four classes were generated with overall accuracy of over 90%. To detect the land cover changes during 4 periods between 1956 and 2006, land cover maps of 1956, 1972, 1990 and 2006 were compared, and change maps and Tables were made. The results showed significant urban expansion, vegetated and barren lands losses and stability in rocks and mountainous areas during 1956-2006. Manuscript profile -
Open Access Article
84 - Designing a model for forecasting the return of the stock index (with emphasis on neural network combined models and long-term memory models)
Reza Najarzadeh Mehdi Zolfaghari Samad GolamiThis study presents the new hybrid network of GARCH family and an artificial neural network to predict the Tehran Stock Exchange index during the period of 2008-2017. The existence of long-term memory in the conditional variance of the Tehran stock returns causes use in MoreThis study presents the new hybrid network of GARCH family and an artificial neural network to predict the Tehran Stock Exchange index during the period of 2008-2017. The existence of long-term memory in the conditional variance of the Tehran stock returns causes use in addition GARCH and EGARCH models with short- memory, long-term memory models. In addition to long-term memory models, considering the better performance of hybrid models in predicting financial data of the Garch family models (short and long-term) are combined with the artificial neural network. Using hybrid models the return of stock index was forecast for the next 10 days and its accuracy was evaluated using the evaluation criteria. The results showed that the hybrid FIEGARCH with the student-t distribution model was more efficient in forecasting return of stock and had a lower forecast error than others models Manuscript profile -
Open Access Article
85 - Designing a hybrid intelligent model for predicting the Financial Richness
fatemeh shahbazadeh ebrahim abbasi Hosein Didehkhani Ali KhozeanThis study aims to present an intelligent model for predicting financial opulence in the security companies as a system that supports the decisions. For this reason, by investigating background of the seventeen numbers of variables as a predictor variable for predicting MoreThis study aims to present an intelligent model for predicting financial opulence in the security companies as a system that supports the decisions. For this reason, by investigating background of the seventeen numbers of variables as a predictor variable for predicting the class of financial opulence from valid sources of central security Site G.A.A during the years 1390- 1395 had been extracted. For conducting this investigation, there had been used of the data of Security Industry, during the years 1390 to 1395. In this investigation, first, we compare the results of applying different models of prediction based on Data Mining and in the second stage, we investigate the ranking of predicting algorithms. The finding results of this investigation showed that the financial opulence with acceptable precision is predictable and the extracted model by using the decision tree has very high precision and capability. Manuscript profile -
Open Access Article
86 - Provide intelligent classification model based on perceptron artificial neural network (MLP) and hierarchical analysis (AHP) in digital marketing services to prioritize liquidity and investment risk
Alireza Ashouri Roudposhti Hormoz Mehrani Karim HamdiThe present study, using machine learning and polling techniques, attempts to examine the automated strategic model in order to classify and explore the ideas presented about specific services that have been studied in this area in the field of investment. Provide resul MoreThe present study, using machine learning and polling techniques, attempts to examine the automated strategic model in order to classify and explore the ideas presented about specific services that have been studied in this area in the field of investment. Provide results in digital marketing services. The neural network-based model, by identifying related opinions, measures different characteristics at different levels of evaluation and automatically categorizes opinions depending on the quality of the presentation. Financial crises in the banking system are usually due to the inability to manage financial risks and liquidity, which is a factor in the lack of transparency and ability to manage capital. Thus, the existence of such uncertainties has reduced the interest of investors in industrial and executive partnerships. This article has been established with the aim of identifying the factors affecting liquidity risk and also providing an intelligent model for predicting and classifying liquidity risk factors, identifying and prioritizing the factors involved. For this purpose, the method of intelligent measurement using perceptron neural network (MLP) has been used, which is considered as a practical approach to artificial intelligence. For this purpose, the necessary studies on financial information and liquidity in Bank Mellat branches in Tehran (consisting of 36 branches) have been considered and for the sample population, a random cluster set of 374 selected customers and investors has been used. Manuscript profile -
Open Access Article
87 - Developing a model for predicting the Tehran Stock Exchange index using a combination of artificial neural network and Markov hidden model
Leila Talaie Kakolaki Mehdi Madanchi Taghi Torabi Farhad GhaffariThe purpose of this study was to design a new model for predicting the Tehran Stock Exchange index using pattern recognition in a combination of hidden Markov model and artificial intelligence. The present study is an applied type and mathematical analytical method. Its MoreThe purpose of this study was to design a new model for predicting the Tehran Stock Exchange index using pattern recognition in a combination of hidden Markov model and artificial intelligence. The present study is an applied type and mathematical analytical method. Its location is the Tehran Stock Exchange and during the years 2010 to 2020. Findings showed that the prediction error rate with artificial neural network has a higher accuracy than Markov's hidden model. Also, the prediction error of the hybrid model is much lower than the other two models for predicting the total stock index of Tehran Stock Exchange, so it has higher accuracy for forecasting stocks. According to the MAPE index, the hybrid model method could improve the predictive power of the artificial neural network by 0.044% and also improve the predictive power of the hidden Markov model by 0.70%. Manuscript profile -
Open Access Article
88 - Presenting the developed model of Benish by using tunneling phenomena based on artificial neural network technique and particle swarm optimization algorithm to identifying profit manipulating companies
Farhad Azadi Mehrdad GhanbarI Babak Jamshidi navid Javad MasodiToday, profit rates and the possibility of managing and manipulating the profits are clear to all, and researchers have always sought solutions to remove the uncertainties facing investors and stakeholders when making their financial decisions. To clarify users' decisio MoreToday, profit rates and the possibility of managing and manipulating the profits are clear to all, and researchers have always sought solutions to remove the uncertainties facing investors and stakeholders when making their financial decisions. To clarify users' decision path of financial data users, Beneish (1999) has developed a profit-management predicting model that has yielded different results in different societies. Thus, this article aims to optimize and localize Beneish’s model by adding the Tunneling variable to Beneish’s variable and using a modern neural network and particle swarm algorithms. The statistical research population consisted of 196 companies listed at the Tehran Stocks Exchange from 2014 to 2019. The research method was a descriptive-library method in which the variables are interrelated through the causal-correlational method. From an objective point of view, it is an Ex-Post Facto research design. To analyze the data, the regression method and artificial neural and the PSO algorithms were used. The model analysis results suggested that all financial ratios had significant effects on Beneish’s profit management, as the Tunneling phenomenon and the financial leverage had the highest and lowest effects on predicting Beneish’s profit management, respectively. Manuscript profile -
Open Access Article
89 - Credit risk management in the banking system - A comparative approach of Data Envelopment Analysis and Neural Network and Logistic Regression
Marziyeh Ebrahimi Shghagi Abdollah DaryaborThis research has been done with the aim of identification of effective factors which influence credit risk and designing model for estimating credit Rating of the companies which have borrowed from a commercial Bank in the one-year period by using Data Envelopment Anal MoreThis research has been done with the aim of identification of effective factors which influence credit risk and designing model for estimating credit Rating of the companies which have borrowed from a commercial Bank in the one-year period by using Data Envelopment Analysis and neural network model and comparison of these two models . For this purpose the necessary sample data on financial and non-financial information of 146 companies (as random simple) was selected. In this research, 27 explanatory variables (include financial and non-financial variables) were obtained, by application of factor analysis and Delphi method for examination. Finally 8 variables which had significant effect on credit risk were selected and entered to DEA model. Efficiency of companies was calculated with these variables. Also variables as well as the input vector three-layer perceptron neural network models were added to the model .finally data was processes with logistic regression. Results from data envelopment analysis model and Neural network and Logistic regression in comparisons to the actual results obtained from neural network models to predict credit risk legal customers and credit rating suggest that neural network is more efficient than data envelopment analysis and logistic regression. Manuscript profile -
Open Access Article
90 - Credit rating of manufacturing corporations in Tehran stock exchange withmulti-criteriadecision-makingandartificial neural network models
Maghsoud Amiri Morteza Bakyhoskoie Mehdi Biglari KamiThis paper is investigated the credit rating of manufacturing firms in Tehran Stock Exchange. In this regard, have been extracted financial ratios of public stock companies during the three years from the financial statements. This financial ratios indicates MoreThis paper is investigated the credit rating of manufacturing firms in Tehran Stock Exchange. In this regard, have been extracted financial ratios of public stock companies during the three years from the financial statements. This financial ratios indicates the ability to pay principal and interest of loan. Initially, 50 companies were selected and ranked by the TOPSIS method. Financial ratios are as a criterion and weight of the each criterion are determined by using Shannon entropy method. Then the ranking, companies are classified into four categories. The artificial neural network is trained to classify and after training the neural network are tested. Statistical results show robust classification of neural network. Then all the companies included in this study are classified by neural network. Manuscript profile -
Open Access Article
91 - Presenting of High-frequency Trading System
Mohsen Dastpak Mohammadali RastgavIn emerging markets such as Tehran Stock Market, there is a gap between signals of changing the trend and the beginning of the movements which we can make profit by using a well-designed Algorithmic Trading System. Proposing a high-frequency trading system has advantage MoreIn emerging markets such as Tehran Stock Market, there is a gap between signals of changing the trend and the beginning of the movements which we can make profit by using a well-designed Algorithmic Trading System. Proposing a high-frequency trading system has advantages (taking advantage of intraday stock market volatility) and disadvantages (high amounts of transaction cost due to the high transaction volume) thus we can augment advantages and cotrol the disadvantages by designing the system elaborately. In this research, the “Local Traders” approach for predicting the future trend of stock has been utilized. According to the “Local Traders” approach, there is a local trader (an agent) for each stock which is expert on it. It predicts the future trend of its own stock based on stock’s intraday data and their technical indicators by determining how much it is good to buy, sell or hold. Results show that, the proposed model outperforms the Buy and Hold strategy in all kinds of markets (Normal, Ascending, Descending) even if there is no discount on Transaction Costs. Manuscript profile -
Open Access Article
92 - Evaluate the Ability of Social Networks to Predict the Direction and Stock Prices in Tehran Stock Exchange
Reza Raie Seyed Farhang Hoseini Maedeh Kiani HarcheganiExamining the ability and efficiency of social network on price and direction of stock price is important, because of social network boom. In this research, we observe the herding behavior based on buy and sell offer in one of the Iranian social network (sahamyab.com) u MoreExamining the ability and efficiency of social network on price and direction of stock price is important, because of social network boom. In this research, we observe the herding behavior based on buy and sell offer in one of the Iranian social network (sahamyab.com) using neural network. The duration of research between July 2013to June 2014(1year) and based on TSE is divided to period of bull and bear market. The sample is selected on two hypotheses, ten symbols from active stocks listed by TSE and another ten symbol from most viewed and active on social network. This research done on two parts: direction forecast and price forecast. Historical price and buy/sell offer in social network with 3 to 10 lags used. Feed forward neural network (FFNN) with 3 to 10 data lags and 1 hidden layer and up to twenty neuron used to find optimal network and used to forecast. In price forecast, there is no significant difference, But in directional of stock price forecast, we found that it's significant for most viewed share in bull market and for active share in bear market. Manuscript profile -
Open Access Article
93 - A novel approach to fulfill active portfolio management and automatic stock trading based on feature selection algorithm
Reza Tehrani Mohammad Hendijanizadeh Eisa Noruzian LakvanThis paper intends to present an integrated method to fulfill active portfolio management approach that is based on predicted prices of each six- chosen stock in a four-year time period. First of all Markowitz model is implemented to achieve weight of each stock in each MoreThis paper intends to present an integrated method to fulfill active portfolio management approach that is based on predicted prices of each six- chosen stock in a four-year time period. First of all Markowitz model is implemented to achieve weight of each stock in each year of four year period. Then, twenty two technical indicators as the features of each stock were are regarded as the input data of genetic algorithm (GA) known as a feature selection technique. Two well-performed forecasting methods called k-nearest neighborhood (kNN) and artificial neural network (ANN) are integrated with GA to extract predicted prices of each stock in defined time period. According to predicted price resulted by GA-NN and GA-kNN, a trading strategy is proposed that has an input signal () which imply to buy, sell and do nothing respectively. Returns of created portfolios were compared with the return of buy and hold strategy as the representative of passive portfolio management approach. The portfolio resulted by GA-NN outperformed two other portfolios for our given data. This conclusion emphasizes on superiority of employing active portfolio management to passive portfolio management in terms of tackling fund management issue for our given data. Manuscript profile -
Open Access Article
94 - Trained Radial Neural Networks Based on Variables of Statistical Models and Their Comparison in Bankruptcy Prediction
Alireza Mehrazin Ahmad Zenedel Mohammad Taghipour Omid ForoutanNowadays artificial neural networks have found a special position among these methods. this study seeks to find a better method of building and training artificial neural networks which leads to more accurate predictions of bankruptcy. Meanwhile, three neural networks o MoreNowadays artificial neural networks have found a special position among these methods. this study seeks to find a better method of building and training artificial neural networks which leads to more accurate predictions of bankruptcy. Meanwhile, three neural networks of radial basis function type were built and trained separately by Altman model (1983), Zmijewski model (1984) and combinatory models’ variables. After evaluating the ability of these three models of bankruptcy prediction, their accuracy has been compared. Generally, this study is based on these hypotheses: First, artificial neural network models can predict bankruptcy using Altman, Zmijewski, and combinatory variables. Second, Type I and Type II error rates are equal in the aforementioned artificial neural network models. Time span of 2004 to 2012 (eight years) has been used to select samples from the listed companies in Tehran Stock Exchange. Results show that all three models have the ability of predicting bankruptcy and the model trained with Altman Model’s variables is more accurate than the other two models in this regard. Manuscript profile -
Open Access Article
95 - Dividend Policy Prediction by Multivariable and Univariate Neural Network Models
Mohsen Hamidian M.B. Mohhamadzadeh Moghadam Sajjad Naghdi Javad EsmaeiliThe topic dividend policy is one of the most leading issues in modern corporate finance affecting the firm value. The results of linear methods and regression could not satisfy researchers in forecasting of financial issues such as dividend policy. In this paper, we pr MoreThe topic dividend policy is one of the most leading issues in modern corporate finance affecting the firm value. The results of linear methods and regression could not satisfy researchers in forecasting of financial issues such as dividend policy. In this paper, we present a comparative analysis of the forecasting accuracy of univariate and multivariate Artificial Neural Network using a sample of 183 companies listed in the Tehran Stock Exchange through for the years 2011_2015. This study shows that the application of the multivariate neural network model results in forecasts that are more accurate than Univariate neural network forecasting models. Our findings show that forecast of a multivariate ANN incorporating Marsh and Merton (1987) variables is more accurate than univariate ANNs. Therefore, based on the results of the study we suggest that shareholders, investors and other stakeholders use multivariate ANNs to predict dividend policy of companies listed in Tehran Stock Exchange. Manuscript profile -
Open Access Article
96 - Estimation of Net Present Value (NPV) in industrial and mine projects using General Regression Neural Network
Hossein Badiei Mahyar Yousefi Taleb PargarIn economic studies of industrial and mine projects and estimation of their net present value (NPV) there are many factors as uncertain variables related to the future. Therefore, such studies should be carried out based on forecasting. To obtain reliable results in the MoreIn economic studies of industrial and mine projects and estimation of their net present value (NPV) there are many factors as uncertain variables related to the future. Therefore, such studies should be carried out based on forecasting. To obtain reliable results in these situations, risk analyses methods under uncertainty are used. One of these uncertainty methods is to employ models to be simulated. In these models, some factors as random variables will be studied for the future. Future prediction of each random variable is assessed considering a probability distribution function. The aim of this research work is economic evaluation of Koohzar gold mine in Torbate-Heydariyeh, Iran for a seven years period using simulation and its application in risk and decision management, and estimating its NPV by applying General Regression Neural Network artificial intelligence method. For this purpose, first, probability distributions of the variables were obtained using information from the variables in previous years. Next, distribution functions of uncertain variables are replaced in appropriate cells of discounted cash flow (DCF) table. Then, random sampling was taken from the probability distributions of uncertain variables as an input of cash flow analysis. In the next stage, based on the simulation technique, probability distribution for NPV variations was obtained as the output in the form of graphs and a function. Considering the output, all of the NPV variations can be forecasted. Then, a general regression neural network was designed using simulated results for NPV prediction using input variables. The results show high reliable capability of general regression neural network in prediction of NPV. Manuscript profile -
Open Access Article
97 - Predicting bankruptcy of companies listed on the Stock Exchange using the artificial neural network
Mohsen Vaez-Ghasemi Saeid Ramezanpour ChardehBeing informed of capital market’s companies financial situation is one of the shareholders and economic analysts’ perturbation. Thus, financial market analysts and researchers were looking for methods to predict capital market’s company’s future MoreBeing informed of capital market’s companies financial situation is one of the shareholders and economic analysts’ perturbation. Thus, financial market analysts and researchers were looking for methods to predict capital market’s company’s future conditions. This research is finding a model to predict bankruptcy of stock exchanges market’s companies with using the artificial neural network. In this research we used Zemijewski financial ratios with one macro – economic variable to predict companies’ bankruptcy. Population of study was selected from the accepted companies in Iran’s stock and exchanges organization. Financial ratios have been extracted from companies’ financial statement in a five years’ period between 2010 and 2014, finally we choose 84 companies that divided to salubrious and bankrupt equal number in each. We used multi-layer perceptron (MLP) with back propagation algorithm to create predictor model and data analysis. The network has been trained once with financial ratios and again with additional macro – economic variable to confirm that the accuracy of network model will increase by additional macro – economic variable. Ultimately the designed model in total mode has 92.95 percent of accuracy and 85 percent correct prediction of bankrupted companies for one year earlier of bankruptcy. Manuscript profile -
Open Access Article
98 - Promotion of Effective Level of Investment Management in Iran Capital Market using Artificial Neural Network and Fuzzy Logic
Hossein Amouzad MahdirajiOne of the most important problems in capital market is allocating financial resources in an optimal fashion. In an effective capital market, from an operational point of view, the capital is allocated for the best investment option. Therefore, in order to establish mor MoreOne of the most important problems in capital market is allocating financial resources in an optimal fashion. In an effective capital market, from an operational point of view, the capital is allocated for the best investment option. Therefore, in order to establish more output, making use of appropriate management tools is a step toward more effective market management of transactions. Regarding backgrounds of applying Artificial Neural Networks and Fuzzy Logic in stocks investment and financial prediction, applying them in selecting an appropriate portfolio can lead to desired results for investors. The major goal of the current research is to achieve an optimal investment portfolio in capital market by applying Artificial Neural Network and Fuzzy Logic. Accompanied by Markowitz Model, models were used which were created through Artificial Neural Network. In order to establish investment portfolio, some of those companies were selected which were active in Tehran stock exchange, and which have had positive efficiency from the year 1386 to 1395. In order to evaluate the suggested portfolios in different conditions, the output of different portfolios based on the monthly and yearly output of the member companies were compared and optimization of suggested portfolios using genetic algorithm were carried out. The study shows that using the Fuzzy models versus mentioned models would provide higher output for the investors. Manuscript profile -
Open Access Article
99 - Day-ahead stock price forecasting using hybrid model
Vahid Vafaei Ghaeini Alimohammad KimiagariForecasting financial markets is an important issue in finance area and research studies. Importance of forecasting on one hand and its complexity, on the other hand, researchers have done much work in this area and proposed many methods. In this research, we propose a MoreForecasting financial markets is an important issue in finance area and research studies. Importance of forecasting on one hand and its complexity, on the other hand, researchers have done much work in this area and proposed many methods. In this research, we propose a hybrid model include wavelet transform, ARMA-EGARCH and NN for day-ahead forecasting of stock market price in different markets. At first WT is used to decompose and reconstruct time series into detailed and approximated parts. And then we used ARMA-EGARCH and NN models respectively for forecasting details and approximate series. In this model we used technical index by approximate part to the improvement of our NN model. Finally, we combine prediction of each model together. For validation, proposed model compare with ANN, ARIMA-GARCH and ARIMA-ANN models for forecasting stocks price in UA and Iran markets. Our results indicate that proposed model has better performance than others model in both markets. Manuscript profile -
Open Access Article
100 - یک روش ترکیببی جدید بر اساس تحلیل پوششی داده ها و شبکه عصبی برای بهینه سازی ارزیابی عملکرد
علی نمکین سید اسماعیل نجفی محمد فلاح مهرداد جوادیدر این مقاله ، یک روش جدید ترکیبی از شبکه های عصبی پرسپترون چند لایه و تحلیل پوششی داده ها ارائه می شود که در آن مقادیر ورودی و خروجی برای تعداد زیادی واحد تصمیم گیرنده به عنوان ورودی های شبکه عصبی تعیین می شود. می توان دید که با بکارگیری شبکه عصبی برای حل مسائل تحلیل پ Moreدر این مقاله ، یک روش جدید ترکیبی از شبکه های عصبی پرسپترون چند لایه و تحلیل پوششی داده ها ارائه می شود که در آن مقادیر ورودی و خروجی برای تعداد زیادی واحد تصمیم گیرنده به عنوان ورودی های شبکه عصبی تعیین می شود. می توان دید که با بکارگیری شبکه عصبی برای حل مسائل تحلیل پوششی داده ها نیاز به حل مدل مورد نظر برای هر واحد تصمیم گیرنده نیست و لذا الگوریتم ارائه شده زمان پردازش و استفاده از حافظه را نسبت به آنچه مورد نیاز روش متعارف در تحلیل پوششی داده ها است، به مقدار زیادی کاهش می دهد.جهت بررسی دقت شبکه ارائه شده، چندمطالعه موردی از جمله مجموعه ای از 500شعبه بانک مورد استفاده قرار می گیرد.نتایج نشان دهنده دقت بالا وزمان محاسباتی کمتر(اعتبارلازم) مدل ترکیبی پیشنهادی است. Manuscript profile -
Open Access Article
101 - حل مسایل کنترل بهینه فازی با استفاده از شبکه عصبی بهبود یافته و اصل مینیمم پونتریاگین
S. Askari S. Abbasbandyدر این مقاله سعی بر آن است که بر اساس قابلیت شبکه عصبی مصنوعی و اصل مینیمم پونتریاگین، یک روش جدید جهت حل مسئله کنترل بهینه فازی ارائه شود.در این مقاله سعی بر آن است که بر اساس قابلیت شبکه عصبی مصنوعی و اصل مینیمم پونتریاگین، یک روش جدید جهت حل مسئله کنترل بهینه فازی ارائه شود. Manuscript profile -
Open Access Article
102 - Modeling of Groundwater Quality Parameters Using Artificial Neural Network and Geostatistics Models (Case Study: Zeidoun plain)
Abdol Amir Echreshzadeh Aslan EgdernezhadBackground and Aim: One of the obstacles to develop sustainable is the poor quality of water. The assessment of water quality is usually based on chemical decomposition and measurement of chemical parameters of water. Measuring these parameters in big area is costly and MoreBackground and Aim: One of the obstacles to develop sustainable is the poor quality of water. The assessment of water quality is usually based on chemical decomposition and measurement of chemical parameters of water. Measuring these parameters in big area is costly and time-consuming, as result it required to estimating methods for prediction of those parameters. The purpose of this study is to model the groundwater quality parameters of Zeydoon plain using ANN+PSO and geostatistics models. Methods: For this purpose, the information of 42 observation wells in Zeidoon plain on a monthly basis for 7 years has been used. Neural network model inputs including qualitative parameters SO42- ، pH ، HCO32-، Na+، Mg2+، Ca2+، TDS، SAR and EC were considered. Findings: The results of simulation of groundwater quality parameters using ANN + PSO model showed that in SAR simulator model the highest simulation accuracy is related to the model with sigmoid logarithm function, in EC simulator model the highest accuracy is similar. The construction is related to the model with the stimulus function of the sigmoid tangent. Also, in the TDS simulator model, the highest simulation accuracy of the model with the sigmoid tangent stimulus function was obtained. As RMSE and MAE have the lowest value and R2 index has the highest value. The results of simulation of groundwater quality parameters using the geostatistical model showed that the highest accuracy of the kriging model in the simulation is related to EC, SAR and TDS parameters, respectively. Discussion and Conclusion: Finally, comparing the results of comparing the results of ANN + PSO model and Kriging model showed that ANN + PSO model is more accurate in simulating groundwater quality parameters of Zidon plain than Kriging model. Also, the results of this research showed that the combination of intelligent models with optimization algorithms with correct architecture and complete model inputs are used as a useful tool for simulating groundwater quality parameters. Manuscript profile -
Open Access Article
103 - Evaluation of the most efficient supervised classification algorithm in monitoring growth changes in Tehran
Aida Ashjaee Seyed Masoud Monavari Jalil Imany Harsini Maryam Robati Zahra AziziBackground and Objectives: The urban sprawl is a dynamic and complex phenomenon, and the most effective factor is land use-cover change Coordinated by with the growth of population and economy, and the resulting changes affect vegetation and the functioning of urban eco MoreBackground and Objectives: The urban sprawl is a dynamic and complex phenomenon, and the most effective factor is land use-cover change Coordinated by with the growth of population and economy, and the resulting changes affect vegetation and the functioning of urban ecosystems. In this paper, identification of the most appropriate classification algorithm to investigate the effect of urban sprawl growth in the east of Tehran city in the time period of 1986 to 2016 on land use-cover changes of Jajroud protected area has been studied. Material and Methodology: In this research, the land cover-use changes map was prepared using the supervised classification method and the comparison of three neural network algorithms, minimum distance and maximum likelihood was done in ENVI 5.3.1 software environment. Findings: Land use-cover changes from 1986 to 2016 (period of 30 years) shows the increase of land use-cover area including compact rangelands 58.45%, arid region 91/19%, urban 65/57%, and forest 74/47%. In 2016 compared to 1986. Discussion and Conclusion: By comparing and examining three supervised classification algorithms including neural network, minimum distance, maximum likelihood, the neural network method has been the most suitable algorithm to identify land use-cover changes. Manuscript profile -
Open Access Article
104 - Evaluating the efficiency of artificial neural network in prediction of Electrical conductivity of Zarrinehroud River
Ali Khoshnazar Touraj Nasrabadi Pouyan Abbasi MaedehSixteen stations on Zarrinehroud River were sampled and parameters like temperature, alkalinity, Ph, electrical conductivity, dissolved oxygen and major anions and cations were measured on water samples. Afterwards, Pearson correlation coefficient between EC and other p MoreSixteen stations on Zarrinehroud River were sampled and parameters like temperature, alkalinity, Ph, electrical conductivity, dissolved oxygen and major anions and cations were measured on water samples. Afterwards, Pearson correlation coefficient between EC and other parameters were determined and the ones with lower cost of measurement were considered as the inputs of neural network models. Finally, the model number 5 with tangent Simulating algorithm and Levenberg-Marquet training Algorithm with minimum prediction error was accepted. The maximum determination coefficient, RMSE and NRMSE Were estimated to be 0.98, 168.33 and 0.28 respectively. Furthermore, it is observed that pH has a remarkable sensitivity more over 60 percent on the artificial neural network prediction. Manuscript profile -
Open Access Article
105 - Modeling quality parameters EC, SAR and TDS in groundwater using artificial neural network (case study: Mehran Plain and DEHLORAN)
Mir Mehrdad Mirsanjari Fatemeh Mohammadyari Reza Basiri Fatemeh Hamidi pourGiven the importance of ground water for drinking and agriculture sector, simulation and forecasting changes its quality is an increasing human needs. In this study, the modeling of water quality parameters TDS and EC based on other chemical components of the major an MoreGiven the importance of ground water for drinking and agriculture sector, simulation and forecasting changes its quality is an increasing human needs. In this study, the modeling of water quality parameters TDS and EC based on other chemical components of the major anions and cations, SAR and pH have been carried out. In addition to modeling the sodium adsorption ratio as the dependent variable parameters latitude, electrical conductivity, total dissolved elements and pH values were used as independent variables. The neural network to predict Marquardt Levenberg- groundwater quality parameters were selected. Results showed that high performance neural network to predict the groundwater quality parameters. High levels of correlation coefficient obtained between the values of parameters modeled closely reflects anticipated the measured data and the ability and accuracy of the relationships between input variables with output. Coefficient of determination of all three elements were modeled in three phases: training, validation and testing is over 90 percent Which indicate acceptable accuracy and good learning neural network and efficient network using the learning algorithm and data provided to the network. The results of great importance for the planning and integrated management of water resources and conservation and better utilization of it is important in the study area. Manuscript profile -
Open Access Article
106 - Parameters of predicted changes in the Electrical Conductivity of groundwater in Tehran city with the help of neural network
Naser Mehrdadi Gholam Reza Nabi Bidhendi Akbar Baghvand Hamid Zare Abyaneh Pouyan Abbasi MaedehIn an attempt to examine the quality of ground water in Tehran with respect to the consumption pattern in the last ten years for 71 examination point, three distinct neural networks of different Electrical Conductivity (EC), input and output parameters were MoreIn an attempt to examine the quality of ground water in Tehran with respect to the consumption pattern in the last ten years for 71 examination point, three distinct neural networks of different Electrical Conductivity (EC), input and output parameters were set out . It is observed that, in order to forecast with a great deal of trial and error, the tangent algorithms with the momentum-training algorithm turns out to be less error. As the number of the input parameters is reduced and the training algorithm is fixed with momentum and the stimulating algorithm gives way to the tangent algorithm, error falls off. Finally, three model with one hidden layer, the momentum training algorithms and the stimulating tangent was constructed. The maximum error occurring implies the maximum determination coefficient of 0.986 that its connected to models 1 and 3. Moreover, in line with the neural network laid out in one layer, the minimum normal root mean square error (NRMSE) is supposed to run out at 0.110 in models 1 and 3. According to lesser input parameter of model number 2 and very close approximation to this two models (1and 3) with maximum determination coefficient of 0.96 and the minimum normal root mean square error (NRMSE) 0.176 can be a very close approximation and can decrease inputs parameters and experience for Measurement of input parameters and the estimate is supposed to be excellently acceptable. As regards the effect of the parameters on the forecast made, the neural network involves the predominance of the two sulphate and chloride ions over the sodium parameter. Manuscript profile -
Open Access Article
107 - Application of Artificial Neural Network and Regression Model to Predict the Phenomenon of Dust in the City of Ahvaz
Nabiollah Hosseini Shahpariyan Mohammad Ali Firozi Seyyed Reza Hosseini KahnojDust is one of the phenomena of destructive climate in the western provinces that causes great damage to the environment and many factors are involved in creating this problem. The aim of this study is to predict the phenomenon of dust in Ahvaz city. In this study, Ahv MoreDust is one of the phenomena of destructive climate in the western provinces that causes great damage to the environment and many factors are involved in creating this problem. The aim of this study is to predict the phenomenon of dust in Ahvaz city. In this study, Ahvaz synoptic data during the years (2000-2010) have been used. These data include mean dew point (in degrees Celsius), mean wind speed in knots, relative humidity in terms of average percentage and average monthly rainfall as input, and data on dusty days as target. Networks were introduced. Then, using causal modeling, the relationships between the variables are extracted and finally, the model is tested by neural network and stepwise regression model. The results confirm the ability of more than 74% of the model used to predict the dust phenomenon in Ahvaz. The regression rate of dust data in a linear combination with the variables entered in the equation is equal to 0.651. Also, the resulting coefficient of determination is equal to 0.424 and the modified coefficient of determination is equal to 0.410; That is, in fact, about 41% of the variance of the dust variable is explained and justified through independent variables. Manuscript profile -
Open Access Article
108 - شبیهسازی اثر باکتریهای محرک رشد بر ذرت با استفاده از شبکه عصبی مصنوعی
علی رضا رضایی معصومه نژاد علی علی غفوریان -
Open Access Article
109 - Designing Optimal Neural Networks Controller to Regulate and Control the Output Voltage of DC-DC Boost Converters
Mohammad Zaraei Majid Moradi Zirkohi Najmeh Cheraghi ShiraziDue to the many applications of DC to DC converters in electronics, regulating their output voltage is very important. In many applications it is necessary to change the DC voltage from one level to another. DC -DC converters are used for this purpose. The conversion of MoreDue to the many applications of DC to DC converters in electronics, regulating their output voltage is very important. In many applications it is necessary to change the DC voltage from one level to another. DC -DC converters are used for this purpose. The conversion of DC voltage from one level to another is done by switching elements such as transistors and diodes. Recently, the control of these converters has found a special place in scientific texts. Therefore, one of the objectives of this paper is to control and regulate the output voltage of the converter. The controller proposed in this paper to control the DC voltage level of the converter output is an optimized neural network controller with an algorithm based on colonial competition. The proposed controller function is that first the neural network is designed according to the expected goals of the system and then it is optimized by determining a suitable multi-objective benchmark function using the network structure optimization algorithm. This improves the performance of the control system. Because the proper selection of design parameters has a great role in the performance of the neural network that plays the role of controller. The proposed neural network function is to apply the appropriate signal transducer (PWM signal) to the switching elements in order to increase the performance. The results compared to the PID controller indicate the superiority of the proposed method. Manuscript profile -
Open Access Article
110 - Detection of spleen from abdominal MRI images using neural networks and watershed conversion
Zohreh Miri Alireza Zolghadr asli Mehran YazdiMRI is one of the most useful imaging techniques today. Abdominal MRI imaging is widely used in medical diagnoses such as tumors, tissue diagnostics, etc. Therefore, fast and appropriate segmentation algorithms play an important role in diagnosing diseases, classifying MoreMRI is one of the most useful imaging techniques today. Abdominal MRI imaging is widely used in medical diagnoses such as tumors, tissue diagnostics, etc. Therefore, fast and appropriate segmentation algorithms play an important role in diagnosing diseases, classifying and quantifying tissue, isolating different elements and diagnosing tumors. In this paper, an automatic spleen separation system from abdominal MRI images is presented, which includes two stages of preprocessing and spleen separation algorithm. Pre-processing is used to de-noise and improve image quality. Isolation of the spleen consists of three stages of segmentation using watershed conversion, calculation of features, and the final step of comparing these features with reference values. Any element whose properties are closer to the reference properties is labeled as a spleen. A forward neural network was used to obtain the reference values, which are the same as the shape of the spleen. The results of the spleen output obtained are compared with the spleen output extracted by a specialist, and the percentage difference between the two outputs is considered as an error. Manuscript profile -
Open Access Article
111 - An Ensemble Classifier Method for Breast Cancer Detection Using Genetic Algorithm and Multistage Adjustment of Weights in the MLP Neural Network
Amin Rezaeipanah S. J. Mirabedini ali mobarakiToday, with the increasing spread of science, the use of decision support systems can be of great help in the therapeutic policies of the Doctor. For this purpose, the use of artificial intelligence systems in predicting and diagnosing breast cancer, which is one of the MoreToday, with the increasing spread of science, the use of decision support systems can be of great help in the therapeutic policies of the Doctor. For this purpose, the use of artificial intelligence systems in predicting and diagnosing breast cancer, which is one of the most common cancers among women, is being considered. In this study, the process of diagnosis of breast cancer is done by using multistage weights in the MLP neural network in two layers. In the first layer, the three classifiers are trained simultaneously on the learning set data. Upon completion of the training, the output of the classifier of the first layer is accumulated together with the learning set data in the new sets. This set is given as an input to the second layer superconductor, and the supra-class mapping maps between the outputs of each of the ordinary classifiers of the first layer with the actual output classes. The three-layer structure of the first layer, as well as the second-layer supraclavicle, is a MLP neural network that optimizes the weights, effective properties and the size of the hidden layer simultaneously using an innovative genetic algorithm. In order to evaluate the accuracy of the proposed model, the Wisconsin database is used, which was created by the FNA test. Experiment results on the WBCD dataset the accuracy is 98.72% for the proposed method, which is relative to GAANN, CAFS algorithms provide better performance. Manuscript profile -
Open Access Article
112 - Short-Term Load Forecasting using an Ensemble of Artificial Neural Networks: Chaharmahal Bakhtiari Case
E. Faraji M. Mirzaeian H. Parvin A. Chamkoorii Majid MohammadpourShort-term load forecasting is very important in electrical marketing. Load forecasting is dependent on climatic condition of every region and the previous structures of electrical consumption in that region; so we have accomplished this through employing climatic data MoreShort-term load forecasting is very important in electrical marketing. Load forecasting is dependent on climatic condition of every region and the previous structures of electrical consumption in that region; so we have accomplished this through employing climatic data (including temperature and pressure) and real load consumption of Chaharmahal Bakhtiari. We have evaluated our method using four machine learning algorithms: artificial neural networks (multilayer perceptron), ensemble of artificial neural networks, support vector machine and ensemble of support vector machine. Experimental results indicates that ensemble of artificial neural networks is superior to the others in the field of load consumption forecasting of Chaharmahal Bakhtiari. Manuscript profile -
Open Access Article
113 - Modeling and quantitative structure-activity study of some carboxylate derivatives as anticancer drugs using multivariate linear regression and artificial neural networks
mehdi nekoei Mahmood Ebrahimi پرستو فتاحی Behzad ChahkandiChemotherapy is one of the most effective treatments for cancers, but many cancers become resistant to the therapeutic effects of a drug during treatment with chemotherapy, which is called Multi Drug Resistance. Currently, some new drugs, including carboxylate derivativ MoreChemotherapy is one of the most effective treatments for cancers, but many cancers become resistant to the therapeutic effects of a drug during treatment with chemotherapy, which is called Multi Drug Resistance. Currently, some new drugs, including carboxylate derivatives, have been used to reduce drug resistance. In the present study, a structure-activity quantitative relationship (QSAR) study was performed to predict the drug activity of some carboxylate derivatives using multivariate linear regression (MLR) and artificial neural networks (ANN). First, the structure of drug compounds, drawing and appropriate group of descriptors were calculated. Then, the step selection method was used to obtain the best descriptors that were most related to the drug activity of the compounds. First, the linear model of multiple linear regression (MLR) was developed. ANN was then used to obtain better results. Statistical data show the superiority of ANN method over MLR method. Manuscript profile -
Open Access Article
114 - Application of multivariate linear regression and artificial neural networks to predict the antimicrobial activity of some anilide derivatives by quantitative structure-activity relationship (QSAR) method
mehdi nekoei Parastoo pourali Syed Hamed MppsaviQuantitative structure-activity relationship (QSAR) study was performed to predict the antimicrobial activity of some anilide derivatives using multivariate linear regression (MLR) and artificial neural networks (ANN). First, the structure of the compounds, the drawing MoreQuantitative structure-activity relationship (QSAR) study was performed to predict the antimicrobial activity of some anilide derivatives using multivariate linear regression (MLR) and artificial neural networks (ANN). First, the structure of the compounds, the drawing and the appropriate group of descriptors were calculated. Then, the step selection method was used to obtain the best descriptors that were most associated with the antimicrobial activity of the compounds. With this method, 5 descriptors were selected and first the linear MLR model was constructed. Then, artificial neural network was used to obtain better results. The values of coefficient of determination (R2) and root mean square error (RMSE) for the test series were 0.07 and 0.073 for the MLR linear model and 0.613 and 0.021 for the nonlinear ANN model, respectively. Statistical data show the superiority of ANN method over MLR method. Manuscript profile -
Open Access Article
115 - Modeling and quantitative structure-property relationship studying to predict the half-life of polychlorinated biphenyls using multivariate linear regression and artificial neural networks
سکینه بهرامی نسب مهدی نکوئی سیدعباس طاهریQuantitative structure-property relationship (QSPR) study was performed to predict the half-life of some polychlorinated biphenyl derivatives using multivariate linear regression (MLR) and artificial neural networks (ANN). First, the structure of the compounds, the draw MoreQuantitative structure-property relationship (QSPR) study was performed to predict the half-life of some polychlorinated biphenyl derivatives using multivariate linear regression (MLR) and artificial neural networks (ANN). First, the structure of the compounds, the drawing, and the appropriate group of descriptors were calculated. Then, the step-wise method was used to obtain the best descriptors that were most related to the half-life of the compounds. With this method, 6 descriptors including Lop, GATS5m, GATS8m, LDip, RDF020u, R2v + were selected from the types of topological descriptors, charge, three-dimensional representation of molecules based on electron diffraction and radial distribution function. First, a multiple linear regression linear model was constructed. Then, artificial neural network was used to obtain better results. The values of coefficient of determination (R2) and root mean square error (RMSE) for the test series were equal to 0.716 and 0.050 for the MLR linear model and 0.896 and 0.030 for the nonlinear ANN model, respectively. Statistical data show the superiority of ANN method over MLR method. Manuscript profile -
Open Access Article
116 - Assessment of efficiency of artificial neural network in predicting the trend of desertification processes by using GIS (Case study: Dehloran plain, Ilam)
Soraya Yaghoobi Marzban Faramarzi Haji Karimi Javad SarvarianDesertification is recognized as a main problem in the arid and semi-arid areas. Therefore, identification and prediction of the effective factors in development of desertification are very important for better management of these areas. The main purpose of this study w MoreDesertification is recognized as a main problem in the arid and semi-arid areas. Therefore, identification and prediction of the effective factors in development of desertification are very important for better management of these areas. The main purpose of this study was evaluating the accuracy of an artificial neural network model for predicting the desertification process and selects the most effective criteria on desertification in the Dehloran plain by using the Iranian model for desertification potential assessment (IMDPA). In IMDPA model, water and climatic were selected as effective factors in desertification. In this model, three indicators for climate criteria; annual precipitation, drought index (Standardized precipitation index; SPI and continued drought and for water criteria; ground water table depletion, sodium absorption ratio, Cl, electrical conductivity (EC) and total dissolved solids were evaluated. Each index was rated using of IMDPA model. Then desertification intensity and criteria maps were prepared using a geometric average for predicting period in ArcGIS®9.3. Final data were entered into neural network to predict. The results showed that the neural network model has a high efficiency for predicting the desertification process in the study area. The accuracy of the model was about 80% and mean square error (MSe) was less than one. In addition, the climate factor and the index of EC were found the most effective variables for predicting the desertification process. In 2015-2016 predicted the most important probable criteria affecting the intensity of desertification were climate and water with weighted average 2 (moderate in sub-class1, 2 and 3), 1.84 (moderate in sub-class 1and 2), respectively. Manuscript profile -
Open Access Article
117 - Change detection in the forest cover of Siyahmezgi watershed of Guilan using LandSat images
seyed Armin Hashemi Seyed Raza Fatemi Talab Hamide Kavousi Kalashmi Mortaza Madanipour KermanshahiIn recent decades Caspian forest has been attacked by human intervention. Easy access, abundance and diversity of valuable forest products led to an increase in population density, development of new residential areas and activities of deforestation. Change detection is MoreIn recent decades Caspian forest has been attacked by human intervention. Easy access, abundance and diversity of valuable forest products led to an increase in population density, development of new residential areas and activities of deforestation. Change detection is essential in the assessment and management of natural resources. The aim of this study, was to monitor changes in forests of Siyahmezgi watershed in two time periods (2000 and 2015), using LandSat ETM+ (2000) and OLI (2015) images. Images were geometric corrected using 20 ground control points that are randomly taken from all over the watershed area, and topographic maps. After selection of the best indicators of using Bhattacharyya distance, image classification using an artificial neural network algorithm was performed. The results of classification of neural network method of Siyahmezgi watershed in two time periods (2000 and 2015) showed that overall accuracy is equal to 95.75% and 95.96%, respectively. The area of forest lands during 2000 and 2015 has been reduced in size 213.55 ha. In addition, in this area dense rangelands have declined, but during this period the extent of dry farming and semi-dense rangelands have 169.95 and 9.6 hectares were added, respectively. Manuscript profile -
Open Access Article
118 - Comparing the accuracy of time series classification of Landsat images in monitoring land use change
Ahmad Azimi Najarkolaei Ali Akbar Jamali Zeynolabedin HosseiniIn this research, artificial neural network, maximum likelihood and minimum distance classification methods for analysis of land use changes, during 1989 to 2015, were evaluated and compared images from three Landsat satellite sensors in Sari. After geometric and atmosp MoreIn this research, artificial neural network, maximum likelihood and minimum distance classification methods for analysis of land use changes, during 1989 to 2015, were evaluated and compared images from three Landsat satellite sensors in Sari. After geometric and atmospheric corrections, images of 1989, 2002, and 2015 were categorized under three artificial neural network algorithms, maximum likelihood and minimum distance in five land use classes. After assessing the accuracy of the methods, the Kappa coefficients were calculated for maximum likelihood, artificial neural network and minimum distance of 1989 were 92%, 87% and 65% in 2002, were 89%, 87% and 60%, and in 2015 were 91% %, 90% and 73%, respectively. These coefficients indicate the superiority of the maximum likelihood method in comparison with the other two methods in 1989. Also, the results of land use change over the whole period of the survey (from 1989 to 2015), showed that the areas of residential and irrigated lands were increased by 3615 and 575 hectares, but bare lands, gardens and forests were decreased to 1791, 1127 and 1272 hectares, respectively. According to the results, the two methods of maximum likelihood and neural network were more suitable for land use classification. The maximum likelihood method was better than the neural network method with a difference of 5% in 1989 and 2% in 2002 and 1% in 2015 in the Kappa coefficient. Manuscript profile -
Open Access Article
119 - Forecasts of climate change effects on Amygdalus scoparia potential distribution by using ensemble modeling in Central Zagros
Maryam Haidarian Aghakhani Reza Tamartash Zeinab Jafarian Mostafa Tarkesh Esfahani Mohammad Reza TatianPredicting the potential distribution of plants in response to climate change is essential for their conservation and management. Amygdalus scoparia is a wild almond species native to Iran Therefore, this study aimed at predicting the effect of climate change on the geo MorePredicting the potential distribution of plants in response to climate change is essential for their conservation and management. Amygdalus scoparia is a wild almond species native to Iran Therefore, this study aimed at predicting the effect of climate change on the geographical distribution of A. scoparia in Chaharmahal and Bakhtiari province in the central Zagros region. In this regard, we used 5 modeling approaches, Generalized Linear Model (GLM), Classification Tree Analysis (CTA), Artificial Neural Network (ANN), Generalized Boosting Method (GBM) and Random Forest (RF) to determine relationships between the occurrence of species and environmental factors under the ensemble framework by using Biomod and R software. The results showed that AUC values greater than 0.9 and functioning of all models been excellent. The mean temperature of the driest quarter and Annual precipitation had the most important role for habitat suitability of this species and (85%) changes in A. scoparia distribution was justified. The results of the model showed that 9%, (148680 ha) of in Chaharmahal and Bakhtiari province for the A. scoparia have had high habitat suitability. Area of suitable habitat was calculated by ArcGIS software on current and future climate conditions. Under RCP4.5 and RCP8.5 climate scenario A. scoparia might lose (Respectively 43% and 59%) of its climatically suitable habitats due to climate change factors, by 2050, while in a number of areas (135% and 140%), the current unsuitable habitats may be converted to suitable. The results of this study can be used in planning, conservation and rehabilitation of A. scoparia. Manuscript profile -
Open Access Article
120 - Investigation of the land potential of Kermanshah province for rainfed wheat cultivation using artificial neural network
Milad Bagheri Mohammadreza Jelokhani Noaryki Kayvan BagheriWith increasing population growth and the need for food, wheat as the crop with the largest cultivated area and annual production on a global scale has been especially important. Therefore, identifying and recommending suitable areas for cultivation in each area is esse MoreWith increasing population growth and the need for food, wheat as the crop with the largest cultivated area and annual production on a global scale has been especially important. Therefore, identifying and recommending suitable areas for cultivation in each area is essential. Kermanshah province as the study area is one of the areas that most wheat crops are from among. Therefore, in this study Multilayer Perceptron Neural Network (MLP) with Levenberg-Marquardt algorithm was used to identify the potential of rainfed wheat cultivation. The input layer network consists of 12 layers: land use, average annual rainfall, average rainfall in the autumn, the average spring rainfall, the average annual temperature, average temperatures in spring, average temperatures in autumn, slope, aspect, elevation, humidity the relative and degree of days. The rainfall and temperature layers were prepared using the data from the stations of adventurous and synoptic and the interpolation operation in the ArcGIS environment, respectively. The altitude-related layer was extracted using with a DEM 30×30 meter IRS. To determine the search space of the neural network algorithm, the uncultivated areas are determined and removed from the entire input layers. 210 points of The right place to cultivate were prepared as network training points. Finally, the class of uncultivated areas which 15% and The results of the model consists of five classes: very suitable, suitable, somewhat suitable, poor or very poor, respectively, 5.4, 14.8, 24, 22.5 and 18.3 percent of the total area of the province is allocated. Regression analysis of all data on the network is 91% of the network of the company, effective for the MLP neural network is in these zoning. Manuscript profile -
Open Access Article
121 - Spectral discrimination of important orchard species using hyperspectral indices and artificial intelligence approaches
Mohsen Mirzaie Mozhgan Abbasi Safar Marofi Eisa Solgi Roohollah KarimiStudy spectral reflectance through spectral indices allows the optimal use of the wide range of spectral wavelengths in hyperspectral data. The purpose of this study was to introduce and evaluate the performance of spectral indices to discriminate dominant orchard speci MoreStudy spectral reflectance through spectral indices allows the optimal use of the wide range of spectral wavelengths in hyperspectral data. The purpose of this study was to introduce and evaluate the performance of spectral indices to discriminate dominant orchard species in Chaharmahal Bakhtiari province. In this study, 150 spectral curves were measured in the range of 350 to 2500 mm, from grapes, walnuts and almond trees. After the initial correction, 30 of the most important spectral indices were extracted. Analysis of variance and comparisons of meanings was applied to identify the optimal indices for species discrimination at a 99% confidence level. Then, an artificial neural network (ANN) and support vector machine (SVM) approaches were used to evaluate the performance of indices in species discrimination. ANOVA results indicated that the Moisture Stress Index (MSI), Band ratio at 1,200 nm, normalized phaepophytiniz index (NPQI) and cellulose absorption index (CAI) indices are optimal for discrimination of the studied species. The performance evaluation of the introduced indicators in some of the ANN and SVM enhancement structures has been associated with 100% accuracy in both education and testing, which shows the effectiveness of these studies in distinguishing orchard species. The performance evaluation of the introduced indicators has been validated at 100% in both training and testing stages. This result emphasizes the necessity of performing spectroscopic studies to separate the orchard species before analyzing the hyperspectral images due to their large data volume, high cost and huge data analysis. Manuscript profile -
Open Access Article
122 - The effect of kernel optimization in modeling drought phenomenon using computational intelligence (Case study: Sanandaj)
Jahanbakhsh Mohammadi Alireza Vafaeinezhad Saeed Behzadi Hossein Aghamohammadi Amirhooman HemmasiDrought is one of the most important natural disasters with devastating and harmful effects in various economic, social, and environmental fields. Due to the repetitive behavior of this phenomenon, if the appropriate solutions are not implemented, its destructive effect MoreDrought is one of the most important natural disasters with devastating and harmful effects in various economic, social, and environmental fields. Due to the repetitive behavior of this phenomenon, if the appropriate solutions are not implemented, its destructive effects can remain in the region for years after its occurrence. Most natural disasters, such as floods, earthquakes, hurricanes, and landslides in the short term, can cause severe financial and human damage to society, but droughts are slow-moving and creepy in nature, and their devastating effects appear gradually and over a longer period of time. Therefore, by modeling drought, it is possible to provide plans for drought preparation and reduce the damage caused by it. In this study, computational intelligence algorithms of Multi-Layer Perceptron neural network, Generalized Regression Neural Network, Support Vector Regression with support kernel, and Support Vector regression with the proposed kernel (Support Vector) Regression New kernel has been used to model the drought using the Standardized Precipitation Index. The modeling results, in most cases, showed better performance of the proposed SVR_N model than other models. The values of RMSE and R2 were 0.093 and 0.991, respectively, and the GRNN, MLP, and SVR models performed better in modeling after SVR_N, respectively. Modeling of drought phenomenon in modeling is supported by vector regression method. Manuscript profile -
Open Access Article
123 - Analysis and comparison of land use/land cover changes using artificial neural network (Case study: lands of Taft and Mehriz)
Alireza Sepehri Ali Akbar Jamali Mohamad HasanzadehThe areas of natural resources and vegetation in the Taft and Mehriz townships in recent decades have undergone changes due to its close proximity to the capital of Yazd province. The purpose of this study is to assess the extent and direction and prediction of land and MoreThe areas of natural resources and vegetation in the Taft and Mehriz townships in recent decades have undergone changes due to its close proximity to the capital of Yazd province. The purpose of this study is to assess the extent and direction and prediction of land and vegetation changes in these two cities. In this study, Landsat 5 (1998, 2004 and 2008) and Landsat 8 (2017) satellite images in the period from May to June was used. Modeling land use/land cover changes were carried out based on supervised classification. The process of changes was analyzed using land change modeling and perceptron neural network method. The results showed that 3% (558.8 ha) of land and vegetation cover of Taft to Bayer and 1.3% (209.9 ha) were added to the urban lands and reduced from the lands of gardens by 4.3% (559.2 ha), this is the highest rating. The amount of 2.8% (678.8 ha) of land and vegetation in Mehriz Bayer and 1.7% (184 ha) has been changed to urban residential land. In terms of urban expansion, Mehriz has had the highest amount of reducing vegetation. The results show that land use and land cover changes in Taft city where more than Mehriz city so that most of these changes were related to gardens, but in terms of area of land use and land use conversion in Mehriz city has the highest value More intense. Manuscript profile -
Open Access Article
124 - Modeling and zoning water quality parameters using Sentinel-2 satellite images and computational intelligence (Case study: Karun river)
Kazem Rangzan Mostafa Kabolizade Mohsen Rahshidian Hossein DelfanConsidering the progress made in remote sensing technology, collecting information on the quality of surface water resources by this technology, while reducing the cost and time of traditional sampling, can monitor all surface water zones. In this study, the Sentinel-2 MoreConsidering the progress made in remote sensing technology, collecting information on the quality of surface water resources by this technology, while reducing the cost and time of traditional sampling, can monitor all surface water zones. In this study, the Sentinel-2 satellite images were used to estimate the concentration of acidity, bicarbonate and sulfate parameters. Initially, Sentinel-2 satellite images were pre-processing and then bands and spectral indexes were determined to identify the significant relationship between the parameter values of water quality and images using the multivariate regression method. In the next stage, using Artificial neural network (ANN) and Adaptive Neuro fuzzy inference system (ANFIS) models, the relationship between Sentinel-2 satellite images and water quality parameters were modeled and then their accuracy was calculated for real values. The results showed that in the modeling of sulfate parameter using Sentinel-2 satellite, ANFIS model with relative error equal to 0.0773 and RMSe equal to 0.8014 has a higher accuracy compared to ANN models with relative error equal to 0.1581 and RMSe equal to 1.2477. While, the relative error of the results of the ANN model are obtained 0.0064 and 0.0556 for acidity and bicarbonate parameter, respectively, and RMSe is equal to 0.0702 and 0.2691, respectively. The ANFIS model has a relative error of 0.0165 and 0.0722, and RMSe is 0.1975 and 0.3037 for acidity and bicarbonate parameter, respectively. Finally, using satellite images, the mentioned models were applied to prepare a qualitative map of each parameter along the part of the Karun river. Manuscript profile -
Open Access Article
125 - Study and prediction of land surface temperature changes of Yazd city: assessing the proximity and changes of land cover
Mohammad Mansourmoghaddam Iman Rousta Mohammadsadegh Zamani Mohammad Hossein Mokhtari Mohammad Karimi Firozjaei Seyed Kazem AlavipanahBackground and Objective The expansion of urbanization has increased the scale and intensity of thermal islands in cities. Investigating how cities are affected by these thermal islands plays an important role in the future planning of cities. For this purpose, this stu MoreBackground and Objective The expansion of urbanization has increased the scale and intensity of thermal islands in cities. Investigating how cities are affected by these thermal islands plays an important role in the future planning of cities. For this purpose, this study examines and predicts the effect of land cover (LC) changes in the three classes of LC including urban areas, barren lands, and vegetation on land surface temperature (LST) in the city of Yazd during the last 30 years using Landsat 5 and 8 images. This study also examines the effect of the ratio of proximity to the barren land and vegetation classes during this period to examine how the recorded LST is affected by the mentioned ratio.Materials and Methods The LC maps of Yazd city were extracted using a supervised Artificial Neural Network classifier for 1990, 2000, 2010, and 2020. Terrestrial data, google earth, and ground truth maps were used to derive training data. The LST of Yazd was obtained from the thermal band of Landsat 5 and Landsat 8. After that, the LST was classified into six available classes, including 16-20, 21-25, 26-30, 31-35, 36-40, and 41-46°C which has shown that the four last classes play an important role in LST changes in Yazd city during last 30 years. To evaluate the effects of the proximity of barren land and vegetation LC classes on the LST recorded by the sensor, firstly the proximity ratio was calculated in 5×5 kernels for all image pixels. Then the mean of LST was derived based on this ratio of barren and vegetation lands.Results and Discussion The results of this study showed that in Yazd city, from 1990 to 2020, the area of the urban area has grown 91.5 % (33.6 km2) over the last 30 years. Barren and vegetation land, have negative growth in the area over the same period. From 1990 to 2020, barren lands in Yazd experienced a growth -79.4% (21.3 km2), which the sharp growth of urban areas justifies this negative growth in barren lands. Vegetation classes in Yazd from 1990 to 2020, have experienced a growth -68.5% (12.2 km2). The average ground temperature of this city has been constantly increasing during these 30 years. By 2020, the city of Yazd, reaching an average of 38.1°C compared to 29.2°C in 1990, has experienced a 30.4% increase in its average LST. The temperature classes of this city have also moved towards warmer temperature classes in these 30 years. As the main part of the LST area of Yazd, in 1990, in the first place, the class of 26-30 °C with 47 km2 and at the second place the class of 31-35 °C with 26.4 km2 are classified. In 2000, in a reverse trend, the main LST class was 31-35°C with 52.8 km2 as the first place and the 26-30°C class with 20 km2 as the second place. With an increased class, the LST class of 36-40 °C for both 2010 and 2020 with 40.2 and 63 km2 respectively has been recorded as the largest LST class. The LST class of 31-35 °C has been recorded as the second LST class of both years with 33.2 and 9.7 km2, respectively. The difference between these two years is in the growth -70.7% (23.5 km2) of the class area of 31-35°C and the increase of 10.3% (0.8 km2) of the hottest class of the statistical period, 41-46°C, in 2020, compared to 2010. The results of this study also showed that the highest average temperature in all year was recorded for barren lands at 37.3°C. Also, a positive correlation (mean correlation 0.95) was shown between the proximity to barren land cover and the mean LST. However, the sharp upward trend of urban areas in the whole statistical period (91.5% with 33.6 km2) as the second class with the highest average LST after the barren lands with an average of 34.1 °C versus a downward trend of 79.4% (21.3 km2) of barren lands has increased the average LST over a statistical period of 30 years. It is because the decrease of 68.5% (12.2 km2) of vegetation areas as an LC class with the lowest average LST (32.2°C) in the same period, neutralized the effect of decreasing barren lands and intensified the trend of increasing the LST. Meanwhile, a negative correlation (mean correlation -0.97) was established between the ratio of proximity to vegetation and the average LST. The results of forecasting land cover changes in 2030 in the city of Yazd indicate that in a process similar to previous periods, the class of urban areas will increase. This growth will not be significant compared to 2020, with 1.6% (1.1 km2). However, a significant decrease in green areas (vegetation) by -19.6% (1.1 km2) in the same period, along with a slight decrease in barren lands -1.8% (0.1 km2) will cause the earth’s surface to become warmer, and the area of LST classes will be increased by the year. Accordingly, the main area of the LST class in 2030 for the city of Yazd, as in 2020, is forecasted 36-40°C with 58.2 km2 (-7.6% growth compared to 2020). But the dramatic growth of the hottest class of LST over the statistical period (41-46°C) with 166.3% (14.3 km2) growth as the second major class of LST in this year (2030), as well as the negative and dramatic growth of the relatively cooler class 31-35°C with -97.9 % (9.5 km2) in this year indicates the warmer ground surface temperature in 2030.Conclusion The results of this study indicate that in 30 years in Yazd city, the decrease in vegetation in the first place, along with the increase in urban areas in the second place, has caused an increase in LST. Thus, the vegetation class reduces the LST due to its cooling effect considering its water content. In this study, it was shown that by taking all factors into account, the reduction of barren lands will lead to a decrease in LST, and also increasing urban areas with a lower impact factor than barren lands will increase the LST. However, the decrease in the area of green lands (vegetation) in recent years, along with the sharp increase in the area of urban areas has caused an increase in LST. Increasing the proximity to vegetation by creating green areas by increasing the ratio of vegetation in the vicinity of different LC and also reducing the area of barren lands, can be a good solution to deal with the impact of urbanization in recent years on ground surface temperature. Manuscript profile -
Open Access Article
126 - Evaluate the accuracy of Unmanned aerial vehicles (UAV) data on the survey of dieback Buxus hyrcana (Case study: Sisangan forest park-Mazandaran)
Mohammadreza Kargar Younes Babaei Amir Eslam BonyadBackground and ObjectiveSisangan forest park is one of the important habitats of Buxus Hyrcana in Iran. Unfortunately, the park has suffered from dieback in recent years, and many Box trees have been destroyed. Monitoring and management of this zone can be effective in MoreBackground and ObjectiveSisangan forest park is one of the important habitats of Buxus Hyrcana in Iran. Unfortunately, the park has suffered from dieback in recent years, and many Box trees have been destroyed. Monitoring and management of this zone can be effective in controlling, protecting, and supporting it. However, due to the destruction of Box trees, on a large scale, it is not possible to accurately estimate the area using the available data. On the other hand, manual measurements are also very time-consuming and tedious. Therefore, a way must be found to do this process accurately and automatically. Unmanned aerial vehicles (UAV) have made this possible by using highly accurate sensors (spatial resolution). Another solution that can be used to automatically separate dieback trees from green trees is to use different classification methods. The aim of this study is to prove the ability of low-cost UAV data with conventional sensors to detect and zoning areas that have suffered Dieback. Since the cost of UAVs with multispectral sensors (red edge band and near infrared) is very high, it should be possible to reduce this cost. Since the cost of UAV with multispectral sensors (red-edge and near-infrared band) is very high, it should be possible to reduce this cost. Materials and Methods Sisangan Forest Park has located 30km to the east of Nowshahr County, Mazandaran province, at latitude 36º33′30″ to 36º35′30″ N, and longitude 51º47′ to 51º49′30″E. This park is both a tourist destination and many important plant species of the country grow in it. One of the most important of these species is the Buxus Hyrcana. But unfortunately, in recent years they have become snag due to pests and insect infestations. Multirotor UAVs have been used in this research. The camera installed on this device is capable of capturing 20 megapixel images. Imaging operations were performed on December 28, 2017, at 10:00 AM, which lasted 45 minutes. The study area was visited for field sampling and its different points were identified in terms of density of snags and preserved Buxus Hyrcana. Then, three circular pieces with a radius of 60 meters and an area of 1.13 hectares were designed in the zone and the density of snag stands and preserved Buxus Hyrcana stands were determined in these three samples. In each plot, 50 training points were recorded in the places where the Buxus Hyrcana stands were located and also 50 points were recorded in the places where the preserved Buxus Hyrcana stands, floor grass cover, and blackberry was located. In this study, in order to evaluate the accuracy of UAV images in identifying and classifying zones covered with Dieback, the smallest Dieback stands with the smallest canopy width were also recorded. Because UAV images require geometric corrections, they were first corrected geometrically and geographically. They were classified with ENVI software. According to the above explanations, 100 points were recorded in each sample plot, 75 of which were monitored for the classification process and 25 of which were used to evaluate the classification accuracy. Three monitored artificial neural network classification algorithms, maximum likelihood and minimum distance were used to classify these images. Finally, after performing each of the classification steps, a low-pass filter with a size of 3 by 3 pixels was used for smoothing the images. Kappa coefficients and overall accuracy indices were also used to evaluate the results. Results and Discussion In this number of sample plots, 579 stands were measured. Buxus Hyrcana was by far the most frequent in the zone. European hornbeam, Parrotia persica, and Oak were in the next ranks, respectively. The results showed that the artificial neural network algorithm had the best results compared to the other two algorithms. But the results of the artificial neural network also fluctuate according to the condition of the sample piece. This algorithm with an overall accuracy of 97.47% and a kappa coefficient of 0.94 had the best results in the separation and detection of the Buxus Hyrcana snags in the sample plot with the dominance of Buxus Hyrcana snags. After the artificial neural network algorithm, the maximum likelihood algorithm showed more favorable results in separating the Buxus Hyrcana snag stands. The minimum distance algorithm showed good results, but it was not as accurate as of the previous two algorithms. All three algorithms showed poorer results in separating the bases in the sample plot with the dominance of live bases in the sample than the other two sample plots. The sample piece with the predominance of live and green bases compared to the other two sample pieces has more phenomena and effects and in terms of image texture, there are many significant differences compared to the other two sample pieces. All three algorithms showed poorer results in separating the stands in the sample plot by dominance the preserved stands in the sample than the other two sample plots. The sample plot with the predominance of preserved stands compared to the other two sample plots has more phenomena and in terms of image texture compared to the other two sample plots has a lot of significant differences. In this sample plot, in addition to the presence of preserved and snag stands, grass cover and blackberry accessions can also be seen. In this study, the results of classification and detection of Buxus Hyrcana snags using an artificial neural network algorithm were much better than the maximum likelihood and minimum distance algorithms. One of the reasons for the better results of the artificial neural network algorithm is its nonlinearity and non-parametricity. But in classification by traditional algorithms such as statistical methods, they have lower accuracy because they have less flexibility. Parametric types of traditional methods, such as the maximum likelihood algorithm, due to depending on Gaussian statistics, if the data are not normal, cannot have the desired accuracy in classifying and separating classes from each other. In traditional algorithms such as maximum likelihood and minimum distance algorithms, training data play a vital role. In these methods, it is assumed that the distribution within the training samples should be normal so that if this condition cannot be met, the classification accuracy will be greatly reduced. While artificial neural network methods operate based on the characteristics and structure of the data itself. Conclusion The results of this study showed that using the data and ordinary images of a low-cost UAV, it is possible to study the condition of Dieback after the outbreak of the disease and determine its area. Despite the high cost of purchasing expensive sensors to monitor vegetation status, these methods presented in this article can be done at a much lower cost. This method can be of great help to the relevant institutions in determining the area of snag coatings. Manuscript profile -
Open Access Article
127 - Drought prediction and modeling by hybrid wavelet method and neural network algorithms
Jahanbakhsh Mohammadi Alireza Vafaeinezhad Saeed Behzadi Hossein Aghamohammadi Amirhooman HemmasiBackground and Objective A drought crisis is a dry period of climate that can occur anywhere globally and with any climate. Although this crisis starts slowly, it can have a serious impact on health, agricultural products, the economy, energy, and the environment for a MoreBackground and Objective A drought crisis is a dry period of climate that can occur anywhere globally and with any climate. Although this crisis starts slowly, it can have a serious impact on health, agricultural products, the economy, energy, and the environment for a long time to come. Drought severely threatens human livelihood and health and increases the risk of various diseases. Therefore, modeling and predicting drought is one of the most important and serious issues in the scientific community. In the past, mathematical and statistical models such as simple regression, Auto-regression (AR), moving average (MA), and ARIMA were used to model the drought. In recent years, machine learning methods and computational intelligence to model and predict drought have been of great interest to scientists. Computational intelligence algorithms that have been previously considered by scientists to model drought include multilayer perceptron neural network, RBF neural network, support vector machine, fuzzy, and ANFIS methods. In this research, the purpose of modeling and predicting drought is by using three neural network algorithms, including multilayer perceptron, RBF neural network, and generalized regression neural. The drought index used in this research is the standardized precipitation index (SPI). In this research, the wavelet technique in combination with artificial neural network algorithms for modeling and predicting drought in 10 synoptic stations in Iran (Abadan, Babolsar, Bandar Abbas, Kerman, Mashhad, Rasht, Saqez, Tehran, Tabriz, and Zahedan) have been used in different climates and with suitable spatial distribution throughout Iran.Materials and Methods This study, initially using monthly precipitation data between 1961 and 2017, SPI drought index in time scales of 3, 6, 12, 18, 24, and 48 months through programming in soft environment MATLAB software implemented. The results of this step were validated using the available scientific software MDM and Drinc. Then, prediction models were designed using the Markov chain. In this study, a total of six computational intelligence models, including three single models of multilayer perceptron neural network (MLP), radial basis function neural network (RBF), and generalized regression neural network (GRNN), and three hybrids wavelet models with these three models (WMLP-WRBF-WGRNN) have been used to model and predict the SPI index in 10 stations of this research. In implementing all these six models, the MATLAB software programming environment has been used. In this study, four types of discrete wavelets were used, including Daubechies, Symlets, Coiflets, and Biorthogonal. Due to the better performance of the Dobbies wavelet, this type of wavelet was used as a final option in the research. In the Daubechies wavelet used between levels 1 to 45, level 3 showed the best performance among different SPI time scales; therefore, the Daubechies level 3 wavelet was used in all hybrid models of this study. After training all six algorithms used, the evaluation criteria of coefficient of determination (R2) and root mean square error (RMSE) was used to measure the difference between actual and estimated values.Results and Discussion The results of this study showed that computational intelligence methods have high accuracy in modeling and predicting the SPI drought index. In the first stage, the results showed that the individual MLP, RBF, and GRNN models, if properly trained, have close results in modeling and predicting the SPI drought index. In the next step, it was observed that the wavelet technique would improve the modeling results. In using the wavelet technique in combination with three single models MLP, RBF, and GRNN, the choice of wavelet type is also more effective in modeling, so in this research, the first of the four types of discrete wavelets Daubechies, Symlet, Qoiflet, and Biorthogonal in combination with Three single models of this research were used and the results of these four types of wavelets showed the relative superiority of the Daubechies wavelet over the other three wavelets. In using the Daubechies wavelet, since this wavelet has 45 times and the choice of order was also effective in modeling, it was observed by testing the wavelet 45 times that the 3rd wavelet, in general, has higher accuracy in all time scales of SPI index, 3, 6, 12, 18, 24 and 48 months and also in all three algorithms MLP, RBF, and GRNN. Therefore, in this research, the third-order Daubechies wavelet was used in all three algorithms of this research, as well as in all time scales. The results showed that combining the wavelet technique with all three models MLP, RBF, and GRNN will improve the results. The research graphs showed that for the quarterly time scale, the values obtained from the single model prediction in MLP and RBF modeling have a somewhat one-month phase difference compared to the hybrid model, while in the GRNN model, this prediction difference is negligible. The modeling results for both single and hybrid modeling modes indicate that there is no phase difference between the single and hybrid modeling methods in time scales of 6, 12, 18, 24, and 48. For the 12- and 24-month time scales, the single GRNN model had more fluctuations and errors in SPI monthly modeling and forecasting, while the hybrid model in these two-time scales had much better behavior in monthly modeling and forecasting. Distribution diagrams of data related to observational SPI of Abadan station showed that the modeling results for single and hybrid modes in 3 and 6-month time scales are less accurate than other time scales and fit line separation, and its uncertainty is higher than others. However, in all neural network models and in all time scales, the hybrid method has shown more accuracy. The numerical results of the study indicate that in all SPIs and stations under study, the differential values of R2 are positive, which indicates higher values of R2 in the hybrid model than in single neural network modeling, which indicates an improvement in hybrid modeling compared to individual models. Also, the differential values of RMSE are negative in all studied models and stations, which indicates that the amount of RMSE in predicting hybrid models is lower than individual neural network models. In the research graphs, it can be seen that the amount of differences in RMSE and R2 indicates a greater difference in time scales 3 and 6 than the time scales 12, 18, 24, and 48, which somehow goes back to the nature of the data of these time scales. The most significant improvement in R2 and RMSE is from the 3-month low to the 48-month high, respectively.Conclusion From the findings of this study, it can be concluded that artificial neural network algorithms are efficient methods for modeling and predicting the SPI drought index. The use of wavelets in all three models of artificial neural networks will also improve the results. It can also be concluded that for better modeling of the SPI drought index, it is necessary to select the optimal wavelet type and order. From the results of this study, it can be concluded that the wavelet technique has a greater impact on the lower time scales, i.e., 3 and 6 months, than the higher scales, i.e., 24 and 48 months. Manuscript profile -
Open Access Article
128 - Comparing artificial neural network, support vector machine and object-based methods in preparation land use/cover mapsusing landSat-8 images
Farnoush Aslami Ardavan Ghorbani Behrouz Sobhani Mohsen PanahandehPreparing the maps of land use/cover for spatial planning and management is essential. Nowadays, satellite images and remote sensing techniques have widespread applications according to their capabilities to produce the updated data and analyze the images in all discipl MorePreparing the maps of land use/cover for spatial planning and management is essential. Nowadays, satellite images and remote sensing techniques have widespread applications according to their capabilities to produce the updated data and analyze the images in all disciplines such as agriculture and natural resources. In the present study, Artificial Neural Network, Support Vector Machines and Object-Based techniques wereutilized for drawing the land use and vegetation maps in Ardabil, Namin, and Nir counties. The images of LandSat-8 Operational Land Imager (OLI) (2013) were usedafter geometric correction and topographic normalization and classified into 9 land use/cover classes including water bodies, irrigated farming, rainfed farming, meadows, outcrops, forests, rangelands, residential and airport areas. After the accuracy assessment, overall accuracy for the produced maps of ANN, Support Vector Machine (SVM) and Object-based (OB) techniques was estimated as 89.91, 85.68 and 94.37%, respectively and Kappa's coefficients were 0.88, 0.82 and 0.93, respectivelyindicating that the object-based method in comparison with two other methods has more advantages;on the other hand, all three methods could provide the desirable accuracy for the land use/covermaps. Overally, three advanced classification methods were examined in the heterogeneous area with elevation changes up to 3600m using the images of new lunched Landsat 8 and the most appropriate land use/cover mapping method was introduced. Manuscript profile -
Open Access Article
129 - Landslide hazard zonation using artificial neural network (Case study: Sepiddasht-Lorestan, Iran)
Syamak Bharvand Salman SooriThis study was carried out to determine the relative hazard zonation of the slope instabilities and landslide occurrence in Sepiddasht, Iran. The method of Artificial Neural Network with the multiple-layer percepteron structure and the back propagation learning algorith MoreThis study was carried out to determine the relative hazard zonation of the slope instabilities and landslide occurrence in Sepiddasht, Iran. The method of Artificial Neural Network with the multiple-layer percepteron structure and the back propagation learning algorithm were used. In order to study the stability of the slopes, the landslides of the region were initially identified and recorded using satellite images of TM and ETM+, aerial images of 1:50,000, and field surveys (year, 2014). The impact of each factor including slope, aspect, land use, elevation, lithology, precipitation, the distance from the fault road and drainage on the slope instabilities was estimated using the ArcGIS®10.1 software via combining the map of the factors influencing the landslide with the landslide distribution map. Then a proper structure (1-13-9) for the landslide hazard zonation of Sepiddasht region was obtained through training the artificial neural network by MATLAB software. Based on the results of the landslide hazard zonation, 0.18, 12.41, 14.09, 29.85, and 43.52 percent of the region were located in very low, low, medium, high, and very high risk classes respectively. Manuscript profile -
Open Access Article
130 - Estimating changes in forest cover in the Rudsar county by using neural network and maximum likelihood methods
Seyed Reza Fatemti Talab Morteza Madanipour Kermanshahi Seyed Armin HashemiThe acquisition of knowledge about the vegetation plays an important role in soil management. However, vegetation estimating in the usual way, including an overall assessment of the vegetation is time consuming and does not also provide accurate enough infor MoreThe acquisition of knowledge about the vegetation plays an important role in soil management. However, vegetation estimating in the usual way, including an overall assessment of the vegetation is time consuming and does not also provide accurate enough information. Therefore, remote sensing technology is a desirable way for reducing time and cost compared to other usual methods. In this study, forest cover maps were prepared using remote sensing techniques and LandSat ETM+ imagery of year 2000 and LandSat 8 of year 2013. The classification of the study area digital images was performed to prepare land use map classification using maximum likelihood and neural network with participation of different bands. The results showed that the best overall accuracy of image classification using neural networks ETM+ in 2000 and LandSat 8 in 2013 was 0.95 and 0.95 respectively. It was also indicated that the kappa coefficient was estimated 0.91 and 0.91 respectively. The overall accuracy of maximum likelihood method of the collected images of 2000 and 2013 was 0.95 and 0.85, but it was 0.86 and 0.84 for Kappa statistics method. The results also showed a 1054.507 and 635.319 hectares decreasing of forest cover using neural network classification and maximum likelihood classification methods respectively. According to classification accuracy and Kappa statistics, it was observed that the accuracy and kappa coefficient of neural network classification was higher than accuracy and the Kappa coefficient of maximum likelihood method. Manuscript profile -
Open Access Article
131 - Land use change modeling using artificial neural network and markov chain (Case study: Middle Coastal of Bushehr Province)
Mehdi Gholamalifard Mohsen Mirzayi Sharif Joorabian ShooshtariCoastal lands of Bushehr Province has a high importance in terms of marine exporting and importing, oil and gas reserves, agriculture, nuclear plant, suitable condition for fishing and tourist attractions. Therefore new desirable methods for monitoring and modelin MoreCoastal lands of Bushehr Province has a high importance in terms of marine exporting and importing, oil and gas reserves, agriculture, nuclear plant, suitable condition for fishing and tourist attractions. Therefore new desirable methods for monitoring and modeling changes are required to be used in these areas. This study was performed with the aimed of monitoring and modeling land use changes using Artificial Neural Network (ANN) and Markov Chain in Land Change Modeler (LCM) in 23 years period (1990-2011). After model accuracy assessment using kappa coefficient, land cover map of the year 2016 was predicted by the 2006-2011 calibration period. The results indicated that two trends include changes from open lands to agricultural and then quitting these agricultural lands have been observed during the study period. Such that, the agricultural area has increased to 19715.76 hectares from 1990 to 2006,but between 2005 to 2011, only 14.48% of agricultural lands has remained unchanged and the large area of those were finally left. In this study, LCM was able to predict 0.76 of changes correctly. So that it was predicted 12000 hectares increasing of extent urban development in the coastal lands of Bushehr Province in 2016. Manuscript profile -
Open Access Article
132 - Provide a new approach to identify and detect credit card fraud using ANN - ICA
Javad Balaee kodehi Mohammad Tahghighi SharabyanIntroduction: The imperialistic competition algorithm is a method in the field of evolutionary computing that deals with finding the optimal answer to various optimization problems. This algorithm provides an algorithm for solving mathematical optimization problems by m MoreIntroduction: The imperialistic competition algorithm is a method in the field of evolutionary computing that deals with finding the optimal answer to various optimization problems. This algorithm provides an algorithm for solving mathematical optimization problems by mathematical modeling the socio-political evolution process. The imperialistic competition algorithm forms an initial set of possible answers. These initial answers are known as chromosomes in the genetic algorithm, particles in the particle swarm algorithm, and countries in the imperialistic competition algorithm. The imperialistic competition algorithm gradually improves these initial solutions (countries) with a special process that follows and finally provides the appropriate solution to the optimization problem. By imitating the process of the social, economic, and political evolution of countries and by mathematically modeling parts of this process, this algorithm presents operators in a regular form as an algorithm that can help solve complex optimization problems. In fact, this algorithm looks at the solutions of the optimization problem in the form of countries and tries to gradually improve these solutions during an iterative process and finally reach the optimal solution of the problem.Method: The proposed algorithm of this article (combined algorithm of neural network and colonial competition) has used the social-political process of the imperialistic competition algorithm with mathematical modeling in order to provide a strong and efficient algorithm in the field of diagnosis optimization.Findings: Our experiments proved that neural data classification using the transaction rejection option can lead us to a very low error rate, while we are looking for a very high detection rate. In this study, we reached an accuracy rate of 98.54, which is a higher accuracy rate compared to previous methods.Discussion: In this research, credit card fraud detection has been done with the aim of identifying the fraud rate, increasing the accuracy, and applying the lowest system error rate using neural networks and combining it with the colonial competition algorithm. Also, effective features were extracted in the evaluation of fraud detection. It can be concluded that the proposed classification system can have a very high detection performance in credit card financial transactions. Manuscript profile -
Open Access Article
133 - Presenting A Hybrid Method of Deep Neural Networks to Prevent Intrusion in Computer Networks
Mohsen Roknaldini Erfaneh NorooziIntroduction: Nowadays, computer networks have significant impacts on our daily lives, leading to cybersecurity becoming a crucial area of research. Cybersecurity techniques mainly encompass antivirus software, firewalls, and intrusion detection systems. Intrusion dete MoreIntroduction: Nowadays, computer networks have significant impacts on our daily lives, leading to cybersecurity becoming a crucial area of research. Cybersecurity techniques mainly encompass antivirus software, firewalls, and intrusion detection systems. Intrusion detection system is one of the fundamental security tools in the field of computer networks and systems. The primary goal of an intrusion detection system is to identify and alert about any unauthorized activities, threats, or attacks on a system or network. By analyzing the flow of data and network/system events, the intrusion detection system attempts to identify patterns and indicators related to various attacks and intrusions. Intrusion detection systems can operate based on rules or learning. In the rule-based approach, algorithms and rules created by security experts and analysts are used to detect patterns and identify attacks. However, in the machine learning approach, machine learning algorithms and deep neural networks are employed to extract patterns and features related to attacks from real data. Method: This study focuses on the examination and presentation of a combined approach using deep neural networks to prevent intrusions in computer networks. The primary objective of this research is to enhance the efficiency of intrusion detection systems. To achieve this goal, a combined approach of deep learning and artificial neural networks is proposed. This approach utilizes deep neural networks to detect more complex features and improves the model's performance. Results: Simulation results demonstrate that deep neural network methods such as MLP, CNN, LSTM, and GRU yield favorable outcomes compared to other single-layer machine learning techniques. In this study, two combined methods, CNN-GRU and CNN-LSTM, were introduced and tested on the KDD CUP'99 dataset for comprehensive analysis and evaluation. Both combined approaches exhibit high accuracy and lower classification errors compared to other introduced methods. Therefore, it can be concluded that the CNN-LSTM combined approach performs well on the KDD CUP'99 dataset. Discussion: Based on the achieved results, the combined CNN-LSTM and CNN-GRU methods offer very good performance with accuracies of 99.95% and 99.92%, respectively, on the KDD CUP'99 dataset. Among these methods, minor differences in the performance of some parameters for classes may exist, yet both approaches remain acceptable. Hence, it can be concluded that the combined CNN-LSTM approach performs well on the KDD CUP'99 dataset. Manuscript profile -
Open Access Article
134 - Applying of Zhang neural network in time-varying nonlinear function optimization
zeinab Mousavi elaahe Karami Kobra GholamiAbstract: Optimization of nonlinear time-varying functions, as a subset of nonlinear programming, has been widely observed in various economic and engineering models. In energy management, one example of optimizing nonlinear functions with time-variable components is th MoreAbstract: Optimization of nonlinear time-varying functions, as a subset of nonlinear programming, has been widely observed in various economic and engineering models. In energy management, one example of optimizing nonlinear functions with time-variable components is the efficient allocation of energy resources and managing changes in demand and supply, leading to increased efficiency and reduced energy waste. In this article, we intend to use Zhang neural networks for optimizing nonlinear functions with time-varying components. By harnessing the parallel processing power of neural networks, Zhang networks search the solution space faster than traditional methods, significantly reducing the required computational time. In this research, the proposed neural network receives data using MATLAB software. The data is first standardized using standard normalization methods. The data is then divided into four stages: training, testing, experimenting and validation which are further evaluated in five phases. The training data is based on the Luenberger-Madala algorithm for the first layer and a linear function for the second layer. Subsequently, the best network structure is considered with the transformation function and the proposed neural network model is tested in five stages. In this paper, we are going to use Zhang's neural networks to optimize time-varying nonlinear functions. In this direction, a general Zhang discretization model with truncation error O (τ^5) has been used and an attempt has been made to study two general five-stage discrete time models of the Zhang neural network and survey about parameter a_1 and expand the optimal step size h. In this research, using MATLAB software, in order to enter the data into the proposed neural network, they are first normalized with the standard normalization method. The desired data in the research were examined and evaluated in four stages, training, testing, second step of testing and validation and in five phases. The training of the data is based on the Lunberg-Maud algorithm model for the first layer and linear function for the second layer. In the following, the best network structure with transformation function is considered and based on the proposed neural network model, it has been tested in five stages. On the other hand, during the past decades, neural network has attracted the attention of researchers due to its good features, including distributed storage, high-speed parallel processing, hardware applications, and superior performance in large-scale online applications. has attracted Some neural networks have been developed to solve nonlinear optimization during recent decades Manuscript profile -
Open Access Article
135 - Unsupervised Domain Adaptation for image classification based on Deep Neural Networks
Amirfarhad Farhadi Mitra Mirzarezaee Arash Sharifi Mohammad TeshnehlabIntroduction: Domain adaptation has become an important issue today. A high percentage of data processing domain adaptation is done with a significant percentage of studies related to deep learning. Traditional methods often ignore the distance between the intra-class i MoreIntroduction: Domain adaptation has become an important issue today. A high percentage of data processing domain adaptation is done with a significant percentage of studies related to deep learning. Traditional methods often ignore the distance between the intra-class in source domain and target domain. As a result, models can be sensitive to outliers and noisy data, additionally increasing the negative transfer in the model. This method applied GAN to extract appropriate features and then used Fuzzy c-means to cluster train datasets in the target domain. Finally, based on the WMMD metric and CNN, the model estimates the final label data. Five real datasets are selected to generate eight transfer tasks. The results show that the superiority of the proposed model lies in transferring more knowledge from the source domain to the target domain.Method: In this approach, firstly based on GAN extracting features from source domains and the target domain (without labels), then label estimation by Fuzzy c-means clustering, finding the center of Fuzzy c-means on target domain data, new data points with labels in target domain as a new input to feature extraction module and regenerate features by GAN based on new pseudo labels. Afterward, we apply WMMD metrics based on CNN to ultimately assign labels for the target domain. Consequently, classification tasks have been done.Results: Empirical results on various benchmark datasets showcase the exceptional performance of the proposed method compared to state-of-the-art DA approaches, validating the proposed Deep-Learning Unsupervised Domain Adaptation approach efficacy. Overall, the approach shows potential for advancing domain adaptation research by offering an efficient and resilient approach for addressing domain shifts in real-world applications. Experimental results on visual object recognition and a digit dataset reveal that the proposed algorithm is robust, flexible, and significantly superior regarding accuracy compared to the baseline DA approaches. Based on the three and combined digit datasets, 1.7% and 2.4% accuracy improvement are achieved, respectively, compared to the best baseline DA approach results.Discussion: In this research, we addressed the challenging issues of outlier and negative transfer in the context of domain adaptation. Despite significant progress in domain adaptation techniques, outliers and negative transfer instances continue to hinder models' generalization performance across different domains. Based on DNNs and the WMMD metric, our proposed method was designed to mitigate these issues and effectively enhance knowledge transfer between domains. Manuscript profile -
Open Access Article
136 - Using the Modified Colonial Competition Algorithm to Increase the Speed and Accuracy of the Intelligent Intrusion Detection System
Mohammad Nazarpour Navid Nezafati Sajjad ShokouhyarIntroduction: In recent decades, rapid development in the world of technology and networks has achieved, also there is a spread of Internet of thing services in all fields over the world. Piracy numbers have increased, also a lot of modern systems were penetrated. Thus MoreIntroduction: In recent decades, rapid development in the world of technology and networks has achieved, also there is a spread of Internet of thing services in all fields over the world. Piracy numbers have increased, also a lot of modern systems were penetrated. Thus the developing information security technologies to detect the new attack become an important requirement.Method: One of the most important information security technologies is an Intrusion Detection System (IDS) that uses machine learning and deep learning techniques to detect anomalies in the network. In all of the information processing systems, detecting cyber-attacks is one of the main challenges and its effects can be blocked or limited by timely detection of attacks. The IoT system is no exception to this phenomenon, and with the high development of this technology and the expansion of its infrastructure, the need for an intelligent intrusion detection system with high accuracy and speed is essential. Neural networks are modern systems and computational methods for machine learning, knowledge representation, and the application of acquired knowledge to maximize the output accuracy of complex systems. Neural networks have already been used to solve many problems related to pattern recognition, data mining, data compression and research is still underway with regards to intrusion detection systems. One of the disadvantages of using training with classical methods in neural networks is getting stuck in local optimal points. In this paper, we use the meta-heuristic algorithm of Imperial competition algorithm (ICA) to train neural networks and show that in the field of intrusion detection in the IoT system, it can show much better accuracy and speed to classical training methods.Results: Results show that our proposed method has 90% accuracy. This method has a better performance in comparison to classical neural network that has 75% accuracy.Discussion: In this article, we will show that the use of imperial competition evolutionary optimization algorithms instead of traditional methods can increase the accuracy of the IDS system. In addition, evolutionary optimization algorithms are zero order and less complicated than gradient methods. Therefore, using this method, in addition to reducing the cost of system implementation, can increase the speed and accuracy of intrusion detection. In addition, from reliability point of view, we will show that the ICA-based systems are more stable in different implementations. Manuscript profile -
Open Access Article
137 - Face Recognition in Images Using Viola_Jones Method and Image Texture Analysis
Mahdi Hariri Narmineh HeydarzadehIntroduction: Face recognition is one of the most important biometric technologies for people identification, also used in access control. Face recognition is one of the important steps before identity recognition. Usually, one method is used to detect the presence of f MoreIntroduction: Face recognition is one of the most important biometric technologies for people identification, also used in access control. Face recognition is one of the important steps before identity recognition. Usually, one method is used to detect the presence of faces in images. Still, in this research, to increase the accuracy of detection, the combination of two methods is used to improve the detection performance: Viola-Jones algorithm and the matching of image components and texture with face and skin components. In the first step, we use the Viola-Jones algorithm to detect the facial features. In the next step, the features of the eye and nose tissues are analyzed with regression neural networks, and facial features are recognized better according to the facial features. In this research, the texture features of the right and left eyes and the nose of the face are used to increase the matching accuracy. We have used the faces of the FDD-Fold dataset to evaluate the proposed method. Comparing the performance of this method with the RCNN deep network method with a much smaller number of training data, we reached an accuracy of 96.36%, more than the deep learning network. This method gives good results in systems with limited computing ability and average amount of data.The face recognition system is one of the biometric identification systems and one of the most important technologies for people identification, which is also used in access control. Face identification is one of the few biometric methods that, with the advantages of high accuracy and low level of human intervention, is used in cases such as information security, law enforcement and monitoring, traffic control, and registration in attendance systems. This method creates more convenience and development with fewer requirements. then, this method has received more attention during the last twenty years.Face detection is a local binary classification problem that shows the presence of faces in the given image using boxes surrounding them. Although the Viola-Jones method is less accurate than modern methods such as convolutional neural networks; Its much lower efficiency and training parameters compared to the millions of parameters of a typical CNN result in faster training, better accuracy with limited data, and its use in devices with limited computing power such as cameras and mobile phones. The innovation of this method is matching the geometric pattern of the edges to identify the presence of the face in the image, along with matching the skin texture. This method seems to be faster and more accurate than the previous ones.Method: In this research, in the first step, we use Viola-Jones, one of the optimal face recognition algorithms in the image, to detect facial components. In the next step, we use the adaptation of the general shape of facial parts such as eyes, and match the textures in the image with the predicted texture for human skin, to improve the recognition performance and increase the recognition accuracy, in such a way that the regression neural networks examine the eye and nose tissue characteristics and according to the characteristics of the facial tissue, the facial components are recognized by the regression neural network. The investigated features in the texture include minimum and maximum color intensity, mean and median, and variance of the image. The data is given to the regression neural network for training. Here Remarkable thing is matching the overall shape of the human head and face, and in the next step matching the overall shape of the facial parts such as the eyes to improve the accuracy of the presented method. We also use the matching of textures in the image with the texture predicted for human skin to further improve the accuracy of the program's performance. Manuscript profile -
Open Access Article
138 - Sensivity analysis of the effective input parameters upon the population flactuation of the sunn pest adult insect using Artificial Neural Network
Zahra Dustiy Naser Moeini naghadeh Abbas Ali Zamani Leila NaderlooThe Sunn pest, Eurygaster integriceps Put. is the most important pest of wheat and barley in Iran. Many studies have shown that various biotic and abiotic environmental factors affect the population of this pest. In this study, the relationship between population densit MoreThe Sunn pest, Eurygaster integriceps Put. is the most important pest of wheat and barley in Iran. Many studies have shown that various biotic and abiotic environmental factors affect the population of this pest. In this study, the relationship between population density of the Sunn pest adult insect with different environmental factors including sampling date, average daily temperature, average relative humidity, wind speed, wind direction, height from sea level and degree-day was investigated. Field data were collected from two wheat farm of one-hectare in the city of Chadegan, Isfahan province. The used network type was multilayer perceptron with back propagation algorithm and the learning algorithm was Levenberg-Markvart. After sensitivity analysis due to the ease of the model and extraction of effectiveness of factors including four factors of sampling date, temperature, humidity and wind speed were selected. The results showed that a neural network with two hidden layer, 7 neuron in the first hidden layer and three neuron in the second hidden layer, as a sigmoid activation function, and a data percentage of 60, 30, 10 for training, testing and validation for prediction of population fluctuation of the Sunn pest adult insect is used (R2= 0.94). Manuscript profile -
Open Access Article
139 - طراحی و ارائه رویکرد ترکیبی نوین در جهت انتخاب و پیشنهاد مکانی و زمانی بر پایه شبکه عصبی کانولوشن
صدف صفوی مهرداد جلالی محبوبه هوشمند -
Open Access Article
140 - رویکرد کاهش ریسک در مؤسسات مالی برای انتخاب متقاضیان تسهیلات با استفاده از یک پایگاه داده استاندارد بر پایه یادگیری عمیق
حامد حدادی سید ناصر رضوی امین بابازاده سنگر -
Open Access Article
141 - Design a method for removing the cuff of polygraphy system in lie detection testusing MLP neural network
Mohammad Amin Younessi Heravi Mahdi AzarnooshPsychophysiologic verity test is a topic which is used widely in the world. Polygraphy system is usedfor recording the Psychophysiologic signals of a person in these tests and a cuff is used for measuringrelative changes of the arterial blood pressure and recording the MorePsychophysiologic verity test is a topic which is used widely in the world. Polygraphy system is usedfor recording the Psychophysiologic signals of a person in these tests and a cuff is used for measuringrelative changes of the arterial blood pressure and recording the velocity and power of the heart pulsein polygraph system. But the problem is that using cuff in various tests brings undesirable conditionsfor the body. The aim of this paper is to present a method for removing cuff in a way that desiredinformation is resulted in another way. This goal is followed by using arterial photoplethysmograhy.In order to do that, a model is presented which uses arterial volume to estimate blood pressure changessignal. This model was identified with MLP neural network. The output of this model is comparedwith blood pressure changes signal in three levels of signal, feature and classification. Reliability ofthe model was evaluated by presenting appropriate assessment criterions. The results were gained withthe 9.8% relative error, 4.4% relative error power and 80% accuracy of lie recognition in signalsassessed for blood pressure changes.According to the findings, the new method introduced in thisstudy has a comparable accuracy of results to the results of former methods while offering a morecomfortable recording and less diagnostic costs. This new method can be suggested for use as a liedetecting system. Manuscript profile -
Open Access Article
142 - Prediction of LD50 for aniline derivatives (Including some herbicide compounds) using computational methods
E. Mohammadinasab M. RezaeiIn recent decades, computational methods with regard to accurate validation parameters for the determination of the physical- chemistry properties of compounds have been considered by many researchers and have been used as an economic and environmental alternative to sa MoreIn recent decades, computational methods with regard to accurate validation parameters for the determination of the physical- chemistry properties of compounds have been considered by many researchers and have been used as an economic and environmental alternative to saving time and eliminating high costs. In this study, the relationship between the logarithmic values of LD50,) log (LD50)(molkg-1) and molecular descriptors has been investigated for 60 types of aniline derivatives(including some herbicides compounds). At first, the structure of the compounds were drawn by Gauss view05 software and optimized using Gaussian 09 software with B3LYP/6-311++G** method, and then were extracted molecular descriptors. Then inappropriate descriptors were eliminated by genetic algorithm method and the best ones were used for multiple linear regression (MLR) and artificial neural networks (ANN) models. The results showed that the ANN method with the lowest error and the highest coefficient of determination was higher than the MLR method to predicting the log(LD50)(molkg-1) of studied aniline derivatives. Manuscript profile -
Open Access Article
143 - The study of relationship between molecular descriptors and LD50 of organophosphate pesticides
E. Mohammadinasab M. kianpourThe organophosphate pesticides are considered as dangerous chemical pesticides for human health. The possibility of absorbing these toxins through the skin is accessible to many researchers who have been studying the toxicity of organophosphate compounds. Experimental m MoreThe organophosphate pesticides are considered as dangerous chemical pesticides for human health. The possibility of absorbing these toxins through the skin is accessible to many researchers who have been studying the toxicity of organophosphate compounds. Experimental methods are time-consuming and high cost, and they come with a lot of dangers. Quantitative structure activity/property relationship studies provide the capability to access data, information and physico-chemical properties of chemical compounds, using the methods and modeling. In this study, the multiple regression linear method and the artificial neural network with multi-layer perceptron (MLP) model were used to investigate the quantitative relationship of LD50 (mgkg-1) toxicity index with some molecular descriptors of some organophosphate compounds. Investigation of correlation coefficients and root mean square errors values of final models in this study showed that ANN method using the MLP model was higher than the MLR method for prediction of LD50(mgkg-1) of organophosphates compounds. Manuscript profile -
Open Access Article
144 - Prediction of LD50 for carboxylic acid derivatives using multiple linear regression and artificial neural networks models
E. Mohammadinasab f. MohammaeiIn this research, Quantitative Structure–Activity Relationship (QSAR) study has been used for prediction of toxicity values of carboxylic acid derivatives. Firstly, the toxicity (LD50) values of data set of studied compounds were taken from the scientific web book MoreIn this research, Quantitative Structure–Activity Relationship (QSAR) study has been used for prediction of toxicity values of carboxylic acid derivatives. Firstly, the toxicity (LD50) values of data set of studied compounds were taken from the scientific web book and the their structures were drawn with the Gauss view 05 program and optimized at Hartree–Fock level of theory and 3-21G basis set by Gaussian 09 software. Then the dragon software was used for the calculation of molecular descriptors. The unsuitable descriptors were deleted with the aid of the genetic algorithm (GA) and backward techniques, and the best descriptors were used for multiple linear regression (MLR) and artificial neural network (ANN) models. The prediction accuracy of the final model was discussed using the statistical parameters. Leave-one-out cross-validation and external test set of the predictive models demonstrated a high-quality correlation between the observed and predicted toxicity values of all, training, test and validation sets in GA-ANN method. The model by ANN algorithm due to the lower error and higher regression coefficients was clearly superior to those models by MLR algorithm. The proposed model may be useful for predicting log LD50 of new compounds of similar class. Manuscript profile -
Open Access Article
145 - Modeling the Amount of Required Energy and Kinetics of Lavender Drying Using Artificial Neural Network
Mohammad Younesi Alamooti Hamid Khafajeh Mohammad ZareinLavender with the scientific name Lavandula stricta Del is a perennial medicinal plant with a height of about half a meter that grows in different regions of Iran. Drying is one of the oldest methods of preserving materials. The use of neural networks can be used in the MoreLavender with the scientific name Lavandula stricta Del is a perennial medicinal plant with a height of about half a meter that grows in different regions of Iran. Drying is one of the oldest methods of preserving materials. The use of neural networks can be used in the design and selection of optimal working conditions and dryer control. In this study, various parameters of drying, evaluation of mathematical models to determine the best model, evaluation of different topologies of MLP artificial neural network to determine the best network for lavender plant with microwave dryer with power range of 100-1000 watts and The frequency of 2450 MHz is provided in four power levels of 300, 500, 700 and 900 watts. MLP artificial neural network was used to predict the relationship between drying kinetic parameters (moisture ratio and drying rate) and efficiency of energy consumption with changes in microwave power consumption using Statistica software. Among the fitted models, the Midili model was chosen as the best model according to R 2, χ 2 and RMSE criteria. Microwave power levels had an effect on drying time, with drying times of 3 minutes for 900 W power and 11 minutes for 300 W power. In order to predict drying kinetic parameters and energy consumption efficiency, MLP network with one input and three outputs was successfully used. The results generally showed that the MLP artificial neural network is a very powerful tool in predicting drying kinetic parameters and energy efficiency of lavender medicinal plant based on microwave power consumption values. Manuscript profile -
Open Access Article
146 - Quantitative structure–activity relationship on a series of imidazole [1, 2-a] pyridinecarboxamide derivatives as anti-tuberculosis agents
Mohsen Nekoeinia Saeed YousefinejadTuberculosis drug resistance is still one of the most important challenges in the treatment of this infectious disease, and therefore the discovery and development of new effective anti-tuberculosis drugs are always of interest to researchers. In this study, Quantitativ MoreTuberculosis drug resistance is still one of the most important challenges in the treatment of this infectious disease, and therefore the discovery and development of new effective anti-tuberculosis drugs are always of interest to researchers. In this study, Quantitative structure – activity relationship (QSAR) analysis was applied on a series of imidazole[1,2-a] pyridinecarboxamide derivatives as anti-tuberculosis agents. The biological activity of the 18 derivatives were estimated by multiple linear regression and artificial neural network approaches. The four molecular descriptors (nCl, MATS8m, BELe4 and GATS8e) were selected by using stepwise multiple linear regression. The best results of artificial neural network were obtained with a 5-5-1 architecture trained with the feed forward backpropagation algorithm. An external test set containing 5 compounds for evaluating the model's predictive ability was used. The results showed that the artificial neural network approach provides better predictive power compared with multiple linear regression. According to the results of this study, electronegativity, atomic masses and molecular geometry have been found to be important factors controlling the anti-tuberculosis activity. Manuscript profile -
Open Access Article
147 - تخمین خشکسالی با نمایههای SPI و EDI با استفاده از شبکه عصبی مصنوعی- فازی بهینه شده با الگوریتم بهینه سازی ازدحام ذرات (مطالعه موردی: استان کهگیلویه و بویراحمد)
مهدی ملک محمودی مهدی کماسی جعفر جعفری اصل سیما اوحدی -
Open Access Article
148 - تهیهی نقشهی حساسیت به وقوع زمین لغزش با استفاده از مدل شبکه عصبی پرسپترون چند لایه از نوع پیشخور پس انتشار (BP)
سید رضا حسین زاده مسعود مینایی حمید نزاد سلیمانی مهوش نداف سنگانی -
Open Access Article
149 - حساسیت به حرکات توده ای خاک با استفاده از شبکه عصبی مصنوعی،منطق فازی و رگرسیون چند متغیره (مطالعه موردی: حوضه گرماب رود ساری)
محمد ابراهیم عفیفی ابوالفضل بهنیافر -
Open Access Article
150 - توسعه همزمان مدل عددی و هوش مصنوعی جهت شبیهسازی جریان بر روی سرریزهای جانبی
عباس پارسائی امیر حمزه حقی آبی شادی نجفیانمدلسازی عددی پدیده های هیدرولیکی از مهمترین بخشهای علم مهندسی هیدرولیک است. مدل سازی عددی معمولاً شامل حل معادلات دیفرانسیل حاکم بر جریان و همچنین تخمین ضرایب موجود در این معادلات می باشد. در مدل سازی عددی هیدرولیک جریان عبوری از بعضی از سازه های مورد استفاده در Moreمدلسازی عددی پدیده های هیدرولیکی از مهمترین بخشهای علم مهندسی هیدرولیک است. مدل سازی عددی معمولاً شامل حل معادلات دیفرانسیل حاکم بر جریان و همچنین تخمین ضرایب موجود در این معادلات می باشد. در مدل سازی عددی هیدرولیک جریان عبوری از بعضی از سازه های مورد استفاده در مهندسی آب، مانند سرریزهای جانبی نیاز به حل همزمان معادلات دیفرانسیل و تخمین ضریب شدت جریان می باشد. سرریزهای جانبی یکی از انواع سرریزهایی است که به طور وسیعی در انواع پروژه های مهندسی آب مورد استفاده قرار می گیرد. معادله حاکم بر هیدرولیک سرریزهای جانبی معادله متغیر مکانی می باشد که یک معادله دیفرانسیل معمولی است که ضریب شدت جریان نیز در آن حضور دارد. مطالعه بر روی مشخصات هیدرولیکی این نوع سازه به دو قسمت برآورد پروفیل سطح آب و تخمین ضریب دبی تقسیم می شود. حل معادله دیفرانسیل هیدرولیک سرریز جانبی منجر به برآورد پروفیل سطح آب عبوری از روی این سازه می شود. برای مدل سازی ضریب شدت جریان نیز نیاز است که از روش های پرقدرتی مانند انواع مدل های هوش مصنوعی استفاده شود. در این تحقیق با توجه به اهمیت موضوع، برای مدل سازی عددی هیدرولیک سرریز جانبی معادله جریان متغیر مکانی با استفاده از روش رانج کوتای مرتبه چهار حل گردیده و برای تخمین ضریب شدت جریان به توسعه مدل هوش مصنوعی شبکه عصبی تطبیقی (ANFIS) پرداخته شده است. خروجی مدل عصبی تطبیقی به عنوان یکی از ورودی های مدل عددی بشمار میرود. نتایج کلی نشان می دهد که مدل نهایی که از ترکیب دو مدل هوش مصنوعی و مدل عددی تشکیل شده است از توانایی بسیار مناسبی جهت شبیه سازی هیدرولیک این سازه برخوردار می باشد. Manuscript profile -
Open Access Article
151 - ارزیابی تصفیهخانه فاضلاب شهر خرمشهر با استفاده از شبکههای عصبی و مصنوعی و ماشینبردار پشتیبان و امکان استفاده از آن برای مصارف کشاورزی
علی ایران فر علیرضا نیکبخت شهبازی رضا جلیل زاده ینگجهدر این تحقیق از شبکه عصبی و مصنوعی و ماشین‎بردار پشتیبان جهت ارزیابی فاضلاب تصفیه‎خانه شهر خرمشهر استفاده شده است. همچنین امکان استفاده فاضلاب این شهر برای مصارف کشاورزی بررسی گردید. در این تحقیق از مقادیر ماهیانه BOD، COD، TS و TSS که مورد نیاز در این پژوهش بود Moreدر این تحقیق از شبکه عصبی و مصنوعی و ماشین‎بردار پشتیبان جهت ارزیابی فاضلاب تصفیه‎خانه شهر خرمشهر استفاده شده است. همچنین امکان استفاده فاضلاب این شهر برای مصارف کشاورزی بررسی گردید. در این تحقیق از مقادیر ماهیانه BOD، COD، TS و TSS که مورد نیاز در این پژوهش بودند مورد استفاده قرار گرفت و همچنین از این مقادیر جهت ارزیابی فاضلاب تصفیه­خانه شهر خرمشهر جهت استفاده در مصارف کشاورزی استفاده گردید. کاربرد مدل شبکه عصبی­مصنوعی، برای پیش­بینی کیفیت پساب خروجی تصفیه­خانه­های فاضلاب نیز امکان پذیر می­باشد. مدل منتخب ANN (LM) از دقت مناسبی در تخمین به‎هنگام BOD5 برخوردار بود. به هر حال این مدل در پیش­بینی مقادیر حدی بیشینه از عملکرد ضعیف­تری برخوردار بود. با استفاده از الگوریتم بهینه­سازی جست‎وجوی شبکه دو مرحله­ای، مقادیر بهینه مشخصه­های مدل SVM یعنی ɛ، C و γ به ترتیب معادل 037/0، 13 و 472/1 به دست آمد. در نهایت با توجه به نتایج به دست­آمده در این تحقیق مدل SVM برای پیش­بینی به‎هنگام BOD5 برای تصفیه­خانه شهر خرمشهر توصیه شد. با توجه به نتایج به‎دست آمده از آنالیز کیفی فاضلاب ورودی پساب خروجی تصفیه شده راندمان حذف BOD5 برابر با 88 درصد، COD برابر با 92 درصد، TDS برابر با 70 درصد و حذف TSS برابر با 27 درصد می­باشد. شوری پساب تصفیه­خانه فاضلاب شهر خرمشهر با مینیم­شوری 208، ماکزیمم 3050 و میانگین 1544 میکروموس بر سانتی­متر اندازه­گیری شد. بنابراین پساب تصفیه­خانه فاضلاب این شهر در گروه C3، آب­های قابل‎قبول قرار دارد. براساس مقدار سدیم پساب خروجی فاضلاب این تصفیه‎خانه برای آبیاری گندم، جو، سویا، انجیر، زیتون، صنوبر و امثال آن‎ها، بر اساس نمودار ویلکاکس هیچ محدودیتی در استفاده از این پساب وجود ندارد. Manuscript profile -
Open Access Article
152 - تعیین کیفیت آب در طول مسیر رودخانه با استفاده از شبکههای عصبی مصنوعی تکاملی (مطالعه موردی رودخانه کارون بازه شهیدعباسپور- عرب اسد)
محمد نیکو مهدی نیکو تیمور بابائی نژاد آزاده امیری قدرت الله رستم پوررودخانهها به عنوان اصلی ترین منبع تأمین کننده نیاز شرب، کشاورزی و صنعت از اهمیت خاصی برخوردار هستند. از طرفی کیفیت آب از لحاظ شرب نیز در بین پارامترهای کیفی مهم ترین متغیر میباشد. لذا بررسی و پیش بینی تغییرات پارامترهای کیفی در طول یک رودخانه، یکی از اهداف مدیران و بر Moreرودخانهها به عنوان اصلی ترین منبع تأمین کننده نیاز شرب، کشاورزی و صنعت از اهمیت خاصی برخوردار هستند. از طرفی کیفیت آب از لحاظ شرب نیز در بین پارامترهای کیفی مهم ترین متغیر میباشد. لذا بررسی و پیش بینی تغییرات پارامترهای کیفی در طول یک رودخانه، یکی از اهداف مدیران و برنامه ریزان منابع آب، میباشد. در این راستا تعداد زیادی مدلهای کیفیت آب، در زمینه مدیریت بهتر برای حفظ کیفیت آب، گسترش یافته است. در این میان مدلهای شبکه عصبی مصنوعی که با الهام از ساختار مغز بشر عمل مینمایند، به عنوان گزینهای برتر، مورد تحقیق و بررسی قرار میگیرد. این تحقیق بر روی رودخانه کارون، بزرگترین رودخانه کشور و با استفاده از پارامترهای اندازه گیری شده در ایستگاههای موجود در طول رودخانه (بازه شهیدعباسپور- عرب اسد) انجام شده است. بدین منظور، دبی، ماه، طول رودخانه و پارامترهدایت الکتریکی اندازه گیری شده در ایستگاههای شهیدعباسپور، پل شالو، گتوندو عرب اسد به عنوان ورودیهای مدل، در نظر گرفته شد. با استفاده از مدل شبکه عصبی، نسبت جذب سدیم (SAR) و کل املاح محلول (TDS) اندازه گیری شده در همان ایستگاهها نیز پیش بینی میگردد. از جمله مواردی که در این تحقیق به عنوان یک روش جدید استفاده شده است،تعیین شاخصهای کیفی آب، در چند ایستگاه به صورت هم زمان میباشد. به منظور بهینه کردن هرکدام ازمدلهای شبکه عصبی مصنوعی، از الگوریتم ژنتیک استفاده گردید. نتایج نشان میدهد که مدلشبکه عصبی مصنوعی انتخاب شده، نسبت به مدلهای آماری رگرسیون غیرخطی از توانایی، انعطاف پذیری و دقت بیشتری در پیش بینی کیفیت آب در رودخانه برخوردار میباشد. Manuscript profile -
Open Access Article
153 - پیش بینی ابعاد آبشستگی در حوضچه ی استغراق سرریزهای سرویس با روشهای هوش مصنوعی.
علی لشکرآرا سارا خرم زادهپیش ‎بینی دقیق ابعاد حفره آبشستگی در پایین دست سازه های هیدرولیکی از جمله سرریزهای جامی‎ شکل، به دلیل پیچیدگی های ناشی از بررسی همه جانبه و همزمان جریان حاوی آب و رسوب و اعمال کلیه متغیرهای مؤثر در پدیده آبشستگی به سادگی میسر نمی باشد. ابعاد حفره آبشستگی اغلب Moreپیش ‎بینی دقیق ابعاد حفره آبشستگی در پایین دست سازه های هیدرولیکی از جمله سرریزهای جامی‎ شکل، به دلیل پیچیدگی های ناشی از بررسی همه جانبه و همزمان جریان حاوی آب و رسوب و اعمال کلیه متغیرهای مؤثر در پدیده آبشستگی به سادگی میسر نمی باشد. ابعاد حفره آبشستگی اغلب با استفاده از معادلات تجربی تعیین می‎گردد که این روابط در محدوده خاصی از داده ها و شرایط آزمایش پاسخگو می‎ باشد. از آنجایی که ساخت مدل فیزیکی مشکلات و محدودیت هایی به همراه دارد و معمولا در تعیین نگاشت میان پارامتر های مؤثر بر آبشستگی نمی‎ توان اثر دقیق همه پارامترها را در نظر گرفت، لذا در مقاله حاضر بهینه یابی ابعاد حفره آبشستگی برای مجموعه ای از مشاهده ها آزمایشگاهی محققان قبلی طراحی شده است. در این تحقیق ازشبکه عصبی مصنوعی و سیستم تطبیقی عصبی- فازی بهره گیری شده و نتایج آن با معادله حاصل از روش رگرسیون غیرخطی بین داده های مشابه و همچنین فرمول های تجربی پیش ‎بینی حداکثر عمق آبشستگی مقایسه شده است. نتایج این تحقیق حاکی از دقت و برتری قابل ملاحظه سیستم تطبیقی عصبی - فازی با حداکثر خطای 2/5 درصد نسبت به نتایج حاصل از مدل شبکه عصبی و معادله رگرسیون غیرخطی و فرمول تجربی با حداکثر خطا به ترتیب 38/10، 42/12 و 05/14 درصد می‎باشد. Manuscript profile -
Open Access Article
154 - پیش بینی و آنالیز حساسیت تبخیر ماهانه از مخزن سد سیاه بیشه با استفاده از شبکههای عصبی مصنوعی در ترکیب با الگوریتم ژنتیک
آزاده محمدیان شوئیلی حسن فتحیان مهدی اسدی لورفرآیند تبخیر، یکی از مؤلفههای اصلی چرخه آب در طبیعت است که نقش اساسی در مطالعات کشاورزی، هیدرولوژی و هواشناسی، بهره برداری از مخازن، طراحی سیستمهای آبیاری و زهکشی، زمان بندی آبیاری و مدیریت منابع آب ایفا میکند. روشهای زیادی از جمله روشهای بیلان آب، تبخیر از تشت و ر Moreفرآیند تبخیر، یکی از مؤلفههای اصلی چرخه آب در طبیعت است که نقش اساسی در مطالعات کشاورزی، هیدرولوژی و هواشناسی، بهره برداری از مخازن، طراحی سیستمهای آبیاری و زهکشی، زمان بندی آبیاری و مدیریت منابع آب ایفا میکند. روشهای زیادی از جمله روشهای بیلان آب، تبخیر از تشت و روشهای تجربی برای تخمین تبخیر از سطح آزاد، ارائه شده است که هر کدام از این روشها، با محدودیت و خطای اندازه گیری توأم میباشد. امروزه تکنیک جدید استفاده از شبکههای عصبی مصنوعی که مبتنی بر هوش مصنوعی میباشد کاربرد گسترده ای در زمینههای مختلف علمی به ویژه مهندسی آب پیدا کرده است. در این تحقیق با استفاده از مدل شبکه عصبی مصنوعی پرسپترون چند لایه(MLP)، شبکه تابع پایه شعاعی (RBF) و شبکه پیش رونده(FF)،میزان تبخیر ماهانه از مخزن سد سیاه بیشه تا 3 ماه آیندهپیش بینی شد. برای تعیین متغیرهای ورودی مؤثر در مدلهای شبکه عصبی مصنوعی و تعداد نرونها در لایه میانی هر یک از مدلها، از قابلیت بهینه سازی الگوریتم ژنتیک استفاده شد. نتایج نشان میدهد که ضریب همبستگی بین مقادیر اندازه گیری شده و محاسبه شده با مدلهای RBF ، MLPو FFدر برآورد و پیش بینی تبخیر ماهانه از مخزن سد سیاه بیشه به ترتیب برابر با 92/0، 90/0 و 88/0 میباشد. بنابراین مدل RBF از دقت بیشتری نسبت به دو مدل MLP وFFدر برآورد و پیش بینی میزان تبخیر ماهانه از مخزن سد، برخوردار میباشد. نتایج حاصل از آنالیز حساسیت نشان میدهد که تبخیر ماهانه از مخزن سد تا 3 ماه آینده به ترتیب نسبت به زمان وقوع تبخیر بر حسب ماه، فشار هوا در سطح زمین در 1 ، 3 و2 ماه قبل، سرعت باد در سطح 1000 میلی بار در 3 و 2 ماه قبل و دمای هوا در سطح 300 میلی بار در زمان حال بیشترین حساسیت را دارد. Manuscript profile -
Open Access Article
155 - پیشبینی ضریب پخش آلودگی در رودخانه ها با استفاده از مدل ترکیبی شبکه عصبی و الگوریتم ژنتیک
عباس پارسایی امیر حمزه حقی آبی امیر مرادی نژادفرایند پخشیدگی آلودگی در رودخانهها بسیار پیچیده است. مدیریت صحیح کیفیت آب رودخانه نیازمند داشتن اطلاع دقیقی از ضریب پخشیدگی آلودگی است. توسعه مدلهای تجاری جهت شبیه سازی عددی انتقال آلودگی در مجاری روباز مستلزم محاسبه ضریب پخشیدگی میباشد. هرچقدر دقت پیش بینی و محاسبه Moreفرایند پخشیدگی آلودگی در رودخانهها بسیار پیچیده است. مدیریت صحیح کیفیت آب رودخانه نیازمند داشتن اطلاع دقیقی از ضریب پخشیدگی آلودگی است. توسعه مدلهای تجاری جهت شبیه سازی عددی انتقال آلودگی در مجاری روباز مستلزم محاسبه ضریب پخشیدگی میباشد. هرچقدر دقت پیش بینی و محاسبه این پارامتر بیشتر باشد دقت و صحت مدلهای مدل سازی نیز بیشتر میشود و درنهایت برنامه ریزی و تصمیم گیری های مدیریتی متناسب، با دقت و اطمینان بیشتری انجام خواهد شد. روشهای فراوانی برای محاسبه ضریب پخشیدگی وجود دارد ازجمله روشهای تجربی، تحلیلی و ریاضی. روشهای تجربی به علت خطای زیاد، مورد توجه قرار نگرفتهاند. روشهای تحلیلی و ریاضی با منظور کردن فرضیات ساده سازی در مراحل مدل سازی، نیز علی رقم پیچیدگی محاسبات، نتایج قابل قبولی را ارائه نداده اند. بنابراین ارزیابی روابط تجربی به توسعه مدل شبکه عصبی چند لایه پرداخته شده است. معادلات تجربی در بهترین حالت دارای دقتی برابر با ( ) که مربوط به فرمول ارائه شده توسط کاشفی پور و توکلی زاده است. برای تخمین دقیق تر ضریب پخشیدگی مدل شبکه عصبی چند لایه توسعه داده شده است. دقت مدل شبکه عصبی در مراحل آموزش و آزمایش به ترتیب برابر با بوده است. درادامه برای افزایش دقت و کارایی مدل شبکه عصبی، بهینه سازی ضرایب وزنی مورد نیاز شبکه عصبی با استفاده از الگوریتم ژنتیک انجام شده است که عملکرد آن در مراحل آموزش و آزمایش به ترتیب برابر با می باشد. نتیجه نهایی نشان می دهد که می توان دقت مدل شبکه عصبی توسعه داده شده را بدون افزایش تعداد سلول و یا تعداد لایه ها، تا حدود 19 درصد افزایش داد. Manuscript profile -
Open Access Article
156 - شبیه سازی عددی انتشار آلودگی در رودخانهها بوسیله توسعه همزمان روش عددی حجم محدود و مدل شبکه عصبی تطبیقی
عباس پارسائی محمد مهدی احمدی کوروش قادریمطالعه بر روی کیفیت آبهای سطحی از اهمیت ویژهای برخوردار است. این موضوع با توجه به اینکه یکی از مهمترین منابع تأمین آب برای مصارف شرب، کشاورزی و صنایع، رودخانهها هستند نیاز به توجه بیشتری دارد. متأسفانه رودخانهها گاهی به عنوان مکانی برای تخلیه فاضلاب در نظر گرفته می Moreمطالعه بر روی کیفیت آبهای سطحی از اهمیت ویژهای برخوردار است. این موضوع با توجه به اینکه یکی از مهمترین منابع تأمین آب برای مصارف شرب، کشاورزی و صنایع، رودخانهها هستند نیاز به توجه بیشتری دارد. متأسفانه رودخانهها گاهی به عنوان مکانی برای تخلیه فاضلاب در نظر گرفته می شوند. به همین دلیل آلودگی رودخانهها به یکی از مهمترین مسائل و مشکلات در محیط زیست تبدیل شده است. معادله حاکم بر انتشار آلودگی در رودخانه ها، معادله انتقال و پخش است که از نوع معادلات دیفرانسیل جزئی میباشد. این معادله از پرکاربرد ترین معادلات در مهندسی سیالات مخصوصا مهندسی آب میباشد و به صورت کلی معادله حرکت نامیده میشود. در توسعه مدل های کامپیوتری جهت شبیه سازی انتشار آلودگی در آبراهه ها علاوه بر حل عددی معادله حرکت، نیاز به پیش بینی ضریب پخشیدگی نیز میباشد. برای محاسبه ضریب پخش فرمول های تجربی فراوانی ارائه شده است که ارزیابی این روابط نشان می دهد اکثر این معادلات دارای دقت مطلوب نمی باشد. به همین جهت استفاده از روش های هوش مصنوعی اجتناب ناپذیر است. در این مقاله برای گسسته سازی معادله حرکت از روش حجم محدود استفاده و برای تخمین ضریب پخشیدگی شبکه عصبی تطبیقی (ANFIS) توسعه داده شده است. نتایج مدل ANFIS توسعه داده شده نشان می دهد که مدل تهیه شده در مراحل مختلف توسعه مانند آموزش و آزمایش برای تخمین ضریب پخشیدگی از توانایی بسیار مناسبی برخوردار می باشد( و ) بعد از توسعه مدل هوش مصنوعی به توسعه کلی مدل کامپیوتری پرداخته شده است دقت مدل کامپیوتری با حل تحلیلی معادله حرکت و همچنین داده های مشاهداتی رودخانه سورن در انگلستان مورد بررسی قرار گرفت. مقایسه نتایج مدل ارائه شده با دادهای اندازه گیری شده رودخانه سورن در ایستگاه های مورد مطالعه به ترتیب دارای دقتی( و و ) می باشد. نتایج کلی نشان می دهد که مدل کامپیوتری توسعه داده شده از توانایی بسیار مناسبی جهت شبیه سازی انتشار نیز آلودگی در رودخانه ها برخوردار میباشد. Manuscript profile -
Open Access Article
157 - Using Data Mining to Predict Bank Customers Churn
parvin najmi abbas rad maryam shoarThe intensity of finding competition in the industrial and economic space and the market move towards a complete competition market has made the inclination of firms to attract more customers and, instead, have increased the tendency to operate in various service and ma MoreThe intensity of finding competition in the industrial and economic space and the market move towards a complete competition market has made the inclination of firms to attract more customers and, instead, have increased the tendency to operate in various service and manufacturing areas. This policy, which is known for increasing the share of wallet, makes it more important to maintain customer relationships and analyze their relationships, and it is necessary to conduct customer behavioral analysis, customer relationship analysis, and customer behavior forecasting. The present research seeks to identify customers who are turning away and anticipates the decline of customers in order to prevent customers from falling. In this regard, the variables associated with the reversal analysis are first identified and then the bank customers are clustered using a neural network and classified into three categories of loyal, regular, and negative clients. With the receipt of the above labels, a backup vector machine has been used to classify and reverse prediction. Based on the results, the proposed method has the ability to predict rotational deviation of up to 80% and, moreover, has a better performance than the classical decision tree. Manuscript profile -
Open Access Article
158 - Design a closed-loop supply chain in uncertainty conditions taking into account the intermediary warehouses (Case Study: khodrang company)
Laila Arab sayyed mohammad reza davoodiThe management of the supply chain is the process of planning, implementing and controlling the flow of raw material, inventory in the course of construction, final products, as well as the flow of related information from the supply of raw materials up to delivery to t MoreThe management of the supply chain is the process of planning, implementing and controlling the flow of raw material, inventory in the course of construction, final products, as well as the flow of related information from the supply of raw materials up to delivery to the final consumer. The purpose of this study is to design a closed-loop supply chain in uncertainty conditions taking into account the intermediary warehouses in the khodrang company. so that its impact on the process of production and distribution, Berlzom recognizes as much of this concept and position as can be found in Enhance the development of khodrang company. In this research after collecting information and consulting with experts of the company, the model was simplified as much as possible without damaging the data principle. Using the nonlinear programming and neural networks and with the software Metalb and Gram coding. The results of this research in a closed environment without the involvement of external variables in the company itself show that the managers of this company have been able to implement the criteria and indicators related to the ring chain and the Nasal demand and product return levels provide the satisfaction of their major customers and suppliers. Manuscript profile -
Open Access Article
159 - Explaining the categories of support vector machine and neural network for Ranking of bank branches
davod khosroanjom mohamamd elyasi behzad keshanchi Bahare Boobanian shovana abdollahiThere is a lot of information in the banking industry that is of particular importance in identifying it. The use of data mining techniques not only improves quality but also leads to competitive advantages and market positioning. By using data mining and in order to an MoreThere is a lot of information in the banking industry that is of particular importance in identifying it. The use of data mining techniques not only improves quality but also leads to competitive advantages and market positioning. By using data mining and in order to analyze patterns and trends, banks can predict the accuracy of how bank branches are ranked. In this paper, the branches of one of the large commercial banks (number of selected branches 1825 branches and the number of features used 57 features) were performed on real data using support vector machine categories and multi layer perceptron neural network. The evaluation results related to the support vector machine showed that this classifier has lower efficiency for the proposed method. However, the use of neural networks and its combination with PCA showed that it has high performance criteria. Values related to efficiency and accuracy were obtained using neural network with very high accuracy. Manuscript profile -
Open Access Article
160 - Using Artificial Neural Network Methodology and Fuzzy Logic to Design an Intelligent Model for Optimizing and Preventive Maintenance in Interaction with Production in the Textile and Clothing Industry
Sayyed Shahram fatemi Mehrdad Javadi Amir Azizi Sayyed Esmail NajafiIn this research, the intelligent model of preventive maintenance and repairs based on artificial neural network methodology - fuzzy logic with the help of artificial intelligence environment of MATLAB software based on the structure of Falcon's five-layer model of arti MoreIn this research, the intelligent model of preventive maintenance and repairs based on artificial neural network methodology - fuzzy logic with the help of artificial intelligence environment of MATLAB software based on the structure of Falcon's five-layer model of artificial neural networks is presented, the research method is based on systems thinking. After determining the most important factors affecting preventive maintenance and repairs with the help of a questionnaire and based on a dataset of 2,000 samples of data and reports of the Director General of Textile and Clothing Industries of the Ministry of Safety during the years 1396 to 1401 (in the form of six and a half years) and validity Data evaluation by the maintenance and repair experts of 240 industrial units, a smart model was designed, which after the implementation of the model in Borujerd textile factories as the place of implementation of the plan can be claimed if (If); Five "technology" factors have values of 0.9129; Good condition (upper bound of good membership function), "Employees" has values of 0.9239; good condition (upper bound of good membership function), "working environment" has values of 0.8859; relatively good (lower limit of the membership function), "quality" has values of 0.9999; Perfect condition (highest function), "strategy" has values of 0.9999; good status (upper limit) in preventive maintenance and repairs, then: the status of the output variable of the research, i.e."Optimization of preventive maintenance and repairs performance (Y)" will be at its fifth level, i.e. very good, equal to 0.882. Manuscript profile -
Open Access Article
161 - A Model to Predict the Profitability of Pharmaceutical Companies Using Artificial Neural Network
Ali Habibi Ghaffar TariThe aim of this paper is to develop a model to predict the profitability of pharmaceutical companies using the artificial neural network. The research method was descriptive-modeling and statistical population included all active pharmaceutical companies of the Stock Ex MoreThe aim of this paper is to develop a model to predict the profitability of pharmaceutical companies using the artificial neural network. The research method was descriptive-modeling and statistical population included all active pharmaceutical companies of the Stock Exchange including 21 companies. Due to the limited population, no sampling was performed and the whole community was surveyed. The documents of pharmaceutical companies were used to data collection and artificial neural networks were used for data analysis. The results showed that the best model to predict the profitability of pharmaceutical companies is obtained by considering the variables of working capital to total assets, cumulative profit (loss) to total assets, earnings before interest and taxes to total assets, the market value of equity to the book value of debts, sales to total assets and liquidity. Manuscript profile -
Open Access Article
162 - Performance prediction of a steam single-effect absorption chiller by the artificial neural network
Farshad Panahizadeh Mahdi Hamzehei Mahmood Farzaneh-GordDepending on the temperature and pressure of the heat source, single-effect absorption chillers are categorized in two types of hot water and steam single-effect chillers. Due to the ability to use the waste steam in oil, gas and petrochemical industries for air conditi MoreDepending on the temperature and pressure of the heat source, single-effect absorption chillers are categorized in two types of hot water and steam single-effect chillers. Due to the ability to use the waste steam in oil, gas and petrochemical industries for air conditioning and process cooling purposes, the steam type chiller is more widely used. In this study, the artificial neural network is exploited in the prediction of the steam single-effect absorption chiller performance since it is faster and has lower computational cost compared to thermodynamic modeling methods. The perceptron multilayer neural network with the error backpropagation algorithm, the hyperbolic tangent excitation function and the Levenberg-Marquardt learning method with 15285 data points and also the mean squared error estimation index are used. Inputs of the artificial neural network are the inlet cooling tower water temperature, inlet chilled water temperature, inlet steam temperature, outlet chilled water temperature and the solution heat exchanger efficiency respectively. Also, outputs of the neural network are the coefficient of performance and thermal energy consumption of the chiller. Results of this study show that the artificial neural network is capable to predict the coefficient of performance and the thermal energy consumed by the single-effect absorption chiller while the values of mean squared error are 3.183×10^(-7) and 7.466×10^(-8) respectively which verify the accuracy of the method proposed here in absorption chiller performance prediction. Manuscript profile -
Open Access Article
163 - Power and weight optimization of spur gears using metaheuristics and finite element method
Mohammad Sadeghi Ali SadollahGearing is one of the most efficient methods of transmitting power from a source to its application with or without change of speed or direction. In this paper, a spur gear model is optimized aiming to maximize its transmission power and minimize its weight. Several des MoreGearing is one of the most efficient methods of transmitting power from a source to its application with or without change of speed or direction. In this paper, a spur gear model is optimized aiming to maximize its transmission power and minimize its weight. Several design variables named as transmitted power, number of pinion teeth, modules, and thickness of gears have been considered during optimization process. For the sake of optimization, two developed metaheuristics named as water cycle and neural net-work algorithms have been examined using MATLAB programming language platform. Besides, obtained optimization results have been validated and analyzed using well-known commercial computer aided engineering software ANSYS. Based on the ob-tained optimization results, optimum design has been found using optimizers and in terms of engineering analysis good agreement has been observed between the applied finite element approach. Manuscript profile -
Open Access Article
164 - Artificial Neural Networks Models for Rate of Penetration Prediction in Rock Drilling
Naser Ebadati‎ Mehrab ‎ AziziBased on field data, there are various methods to reduce the cost of drilling wells. One of these methods is to optimize the drilling parameters to obtain the maximum rate of penetration (ROP). Many parameters affect ROP. The main purpose of this research is the use of MoreBased on field data, there are various methods to reduce the cost of drilling wells. One of these methods is to optimize the drilling parameters to obtain the maximum rate of penetration (ROP). Many parameters affect ROP. The main purpose of this research is the use of smart networks for the penetration rate of drilling, for this purpose, well input data including drilling depth, duration of the drilling operation, speed of rotation of the drill, weight on the drill, weight and volume of drilling mud as input data. And the drilling penetration rate was prepared as output data from one of the fields located in the Persian Gulf. 70% of data is allocated for network training, 15% of data for validation and 15% of data for sensitivity analysis. According to the obtained results, it was found that using this tool, a good relationship with the total regression coefficient (0.96) is obtained for predicting the penetration rate using a neural network. Also, by repeating the calculations in repetition 12, the best value was obtained, which is equal to 14.24. Manuscript profile -
Open Access Article
165 - Artificial Neural Networks Models for Rate of Penetration Prediction in Rock Drilling
naser ebadati Ronak Parvaneh Mehrab AziziBased on field data, there are various methods to reduce the cost of drilling wells. One of these methods is to optimize the drilling parameters to obtain the maximum rate of penetration (ROP). Many parameters affect ROP. The main purpose of this research is the use of MoreBased on field data, there are various methods to reduce the cost of drilling wells. One of these methods is to optimize the drilling parameters to obtain the maximum rate of penetration (ROP). Many parameters affect ROP. The main purpose of this research is the use of smart networks for the penetration rate of drilling, for this purpose, well input data including drilling depth, duration of the drilling operation, speed of rotation of the drill, weight on the drill, weight and volume of drilling mud as input data. And the drilling penetration rate was prepared as output data from one of the fields located in the Persian Gulf. 70% of data is allocated for network training, 15% of data for validation and 15% of data for sensitivity analysis. According to the obtained results, it was found that using this tool, a good relationship with the total regression coefficient (0.96) is obtained for predicting the penetration rate using a neural network. Also, by repeating the calculations in repetition 12, the best value was obtained, which is equal to 14.24 Manuscript profile -
Open Access Article
166 - Prediction of fragmentation by blasting operation in mines- case study: Gol-e-Gohar iron mine
Ahmad Asadi Eman EnayatollahiBlasting is a key element in mining operation that constitutes near 30% of total mining cost. If this process doesnot carry out correctly, it will increase up to 50% with secondary blasting. A proper basis blasting operation notonly can reduce side effect on the environ MoreBlasting is a key element in mining operation that constitutes near 30% of total mining cost. If this process doesnot carry out correctly, it will increase up to 50% with secondary blasting. A proper basis blasting operation notonly can reduce side effect on the environment, but also can get rid of some undesirable consequences such asback break, fly rock and secondary blasting. Concerning the above notations a predetermination of a method toestimate the size of the fragmented rock and scattered fragmentation is so important and the results are sobeneficial. In this study after performing a series of blasting at Gol-e-Gohar iron mine using artificial neuralnetwork some models for predicting fragmentation have been achieved. In order to select the best blastingpattern concerning fragmentation, Tagochi method was utilized. In these series of experiments, fragmentationresults by using these two methods was 57.5 and 60 cm respectively, which are close to fragmentation at themine. Consequently, some environmental problems were solved by using this pattern of blasting at the Gol-eGohar iron mine. Manuscript profile -
Open Access Article
167 - Application of new Artificial Intelligence Methods in Groundwater Quality Assessment (Case Study: Salmas Plain)
Somayeh Emami Yahya ChoopanGiven all the advancements in water resources management engineering, the problem of groundwater quality assessment is the main problem encountered in most plains of Iran. Therefore, managing and monitoring the quality of water resources is very importance. In this stud MoreGiven all the advancements in water resources management engineering, the problem of groundwater quality assessment is the main problem encountered in most plains of Iran. Therefore, managing and monitoring the quality of water resources is very importance. In this study, we tried to predict and estimate the groundwater quality in the Salmas plain using RBF and GFF models. To achieve this aim, groundwater quality data of Salmas plain during 10 years (2001-2011) were used and results were analyzed according to Wilcox, Scholler and Piper standards. 70% of data were used to train the network and 10% of data were used to validate the two models. Therefore, the remaining 20% of available data was used for network testing. The application of appropriate and applicable statistical parameters showed that RBF model with Levenberg-Marquardt training and 4 hidden layers, has high ability to estimate and predict groundwater quality. Also R2= 0.88 and RMSE= 29.71% in this model. Also the results of using different diagrams show that samples have low hardness and corrosion. Most of the data is in the C3S1 class. According to the results, all the water resources of the study area are acceptable for agriculture, drinking and industry, respectively. Manuscript profile -
Open Access Article
168 - Assessment of Adaptive neural fuzzy inference systems and support vector regression in runoff estimation(A case study:Dez Basin)
Ghazaleh Ahmadian Ahmadabad Mahmoud Zakeri Niri Saber Moazami GoudarziEstimation of discharge flow in basin due to impact on water resource management can have an important economic role.In this research several computationals intelligence techniques suchas:ANN,SVR and ANFIS have been used to prediction the runoff dez basin.correlation be MoreEstimation of discharge flow in basin due to impact on water resource management can have an important economic role.In this research several computationals intelligence techniques suchas:ANN,SVR and ANFIS have been used to prediction the runoff dez basin.correlation between stations was investigated and stations of kamandan,zoorabad and daretakht were eliminated due to small correlation with around stations.then due to lack of human intervention with using xlstat software were evaluated trend of stations and were selected stations without trend.Inorder to evaluate the performance of models were used correlation,RMSE and NSE.Results of this research showed that ANFISwith clustering approach gives better estimation than grid partitioning approach.ANN, ANFIS and SVR have agood ability to simulate the flow of dez basin. Manuscript profile -
Open Access Article
169 - Estimation of TBM Utilization with Artificial Neural Networks
Hamid Reza Nejati Morteza AhmadiDetermination of schedule and bid prices for mechanized tunneling needs to estimate of tunneling machine advance rate. Also determination of machine advance rate needs to estimate machine utilization. Although some empirical equations were proposed for this purpose, the MoreDetermination of schedule and bid prices for mechanized tunneling needs to estimate of tunneling machine advance rate. Also determination of machine advance rate needs to estimate machine utilization. Although some empirical equations were proposed for this purpose, these equations don’t have good accuracy. The aim of this study is estimation of TBM utilization with artificial neural networks. For this purpose data set of open TBM tunneling was gathered and a neural network with 5 inputs (in-situ stress, uniaxial compressive strength, disc thrust, joint orientation factor and Q rock mass classification index) and one output (TBM utilization) was designed. Since input parameters have a good relationship with output parameter, neural network estimate TBM utilization with high accuracy. Manuscript profile -
Open Access Article
170 - Predicting distribution pattern of Bemisia tabaci G. ( (Hem.: Aleyrodidae) by Hybrid neural network With Particle Swarm Optimization Algorithm
Alireza Shabaninejad Bahram TafaghodiniyaToday, with the Advance statistical techniques and neural networks, predictive models of distribution was rapidly developed in Ecology. Purpose of this study was predict and Mapping distribution of Bemisia tabaci G. using MLP neural networks combined with Particle Swarm MoreToday, with the Advance statistical techniques and neural networks, predictive models of distribution was rapidly developed in Ecology. Purpose of this study was predict and Mapping distribution of Bemisia tabaci G. using MLP neural networks combined with Particle Swarm Optimization in surface of cucumber field. Population data of pest was obtained in 2017 by sampling in 100 fixed points in a fallow field in Ramhormoz, to evaluate the ability of neural networks combined with Particle Swarm Optimization to predict the distribution used statistical comparison parameters such as mean, variance, statistical distribution and coefficient determination of linear regression among predicted values and actual values. Results showed that in training and test phases of neural network combined Particle Swarm Optimization algorithm, was no significant effect between variance, mean and statistical distribution of actual values and predicted values. Our map showed that patchy pest distribution offers large potential for using site-specific pest control on this field. Manuscript profile -
Open Access Article
171 - Comparison and Prediction of the Experimental Data for Thermal Efficiency of a Double-Pipe Heat Exchanger with Fe3O4 Nanofluid Using Artificial Neural Networks
محمد اختری مجتبی میرزایی داریوش خسروی مهدIn this study, the thermal efficiency of a double-pipe heat exchanger with Fe3O4-water nanofluid in Reynolds numbers between 2000-21000 and volume fractions between (0.1-0.4% v / v) using artificial neural networks and correlation with experimental data has been evaluat MoreIn this study, the thermal efficiency of a double-pipe heat exchanger with Fe3O4-water nanofluid in Reynolds numbers between 2000-21000 and volume fractions between (0.1-0.4% v / v) using artificial neural networks and correlation with experimental data has been evaluated and predicted. Iron oxide nanoparticles were about 20 nm in size. SEM photography of nanoparticles is provided to show the stability and homogeneity of suspension. Different Reynolds numbers and volume fractions of iron oxide nanofluid are used as the training data for ANN. A two-layer feed-forward neural network with back-propagation Levenberg-Marquardt learning algorithm (BP-LM) was used for heat transfer pre-parameters. Moreover, 70% of data were used in training set and 15% of data were used in evaluation set and remaining data were used as test data to prevent preprocess of network and to study the final efficacy of the network. In addition, based on the experimental data and the use of artificial neural network, data predicted by the neural network are in good agreement with experimental data measured by the double-pipe heat exchanger. The overall verification by the mean squared error (MSE) and correlation coefficient (R2) for the thermal efficiency of a double-pipe heat exchanger is 0.0001 and 0.996, respectively, indicating that prediction is successful. Manuscript profile -
Open Access Article
172 - Study on Young's Modulus, Fracture toughness & Energy of Composites Reinforced by ZnO Nanoparticles through Regression Tree, ANN & RSM
abdorreza alavi gharahbagh Ali dadrasi sasan folladpanjehA study on fracture toughness (KIC), fracture energy (GIC) and Young's modulus of styrene acrylonitrile composites by two volume content of 24% and 34% acrylonitrile has been conducted. ZnO nanoparticles were added to composites up to 1 vt%. Volume percent parameter has MoreA study on fracture toughness (KIC), fracture energy (GIC) and Young's modulus of styrene acrylonitrile composites by two volume content of 24% and 34% acrylonitrile has been conducted. ZnO nanoparticles were added to composites up to 1 vt%. Volume percent parameter has been described as a sensational parameter. The experimental results indicate that adding ZnO nanoparticles increase the mechanical properties and in some cases, it decreases them. Also the experimental results and the results of modeling show that the second order response surface method makes the best prediction. Additionally, the best value for Fracture toughness is 2.283 MPa.m1/2 when the volume percent of styrene acrylonitrile is 34% and volume percent of ZnO is 0.1%. Also, the best value for fracture energy is equal to 1101 J/m2, when the volume percent of styrene acrylonitrile is 34% and the volume percent of ZnO is 0.33%. And finally, this method shows that the best value for Young's modulus is 4.281 GPa when the volume percent of styrene acrylonitrile is 31% And volume percent of the particle is 0.5%. Manuscript profile -
Open Access Article
173 - Clearance Prediction of Rotary System with and without Mechanical Diagnosis by Using Artificial Neural Networks and Particle Swarm Optimization
Mojtaba Hasanluلقی تکیه گاه های موتور و یاتاقان ها سیستم را با کوپلینگ 4 نوع مختلف عیب ابتدا با استفاده از روش تبدیل سریع فوریه فرکانس ها و جابجایی های عمودی شفت در محل دو یاتاقان استخراج نموده و سپس اثر لقی تکیه گاه ها را در حالت حضور و عدم حضور عیوب دیگر مورد بررسی قرار میگیرد. حال Moreلقی تکیه گاه های موتور و یاتاقان ها سیستم را با کوپلینگ 4 نوع مختلف عیب ابتدا با استفاده از روش تبدیل سریع فوریه فرکانس ها و جابجایی های عمودی شفت در محل دو یاتاقان استخراج نموده و سپس اثر لقی تکیه گاه ها را در حالت حضور و عدم حضور عیوب دیگر مورد بررسی قرار میگیرد. حال برای دستیابی به یک مدل بهینه از شبکه عصبی بهمراه الگوریتم بهینه سازی ازدحام ذرات تک هدفه استفاده می کنیم بدین صورت که یکبار فرکانس های سیستم معیوب و بدون بعنوان ورودی شبکه عصبی معرفی میگردند و خروجی مطلوب آن فرکانس سیستم در حالتی که سیستم هیچ گونه عیبی ندارد مدلسازی می شود و سپس در مرحله بعد فرآیند قبل جهت مدل سازی بیهنه با شبکه عصبی را با استفاده از جابجایی های معیوب(وروی شبکه عصبی) و جابجایی سیستم (ورودی مطلوب) مورد ارزیابی قرار میگیرد. Manuscript profile -
Open Access Article
174 - Defects Detection of Rotating Machine Using Vibration Analysis and Neural Network
Seyed Majid Ataei ArdestaniThe base of diagnosing the possible defects of a machine is comparing the frequency ‎spectra of the vibrations at different points with the existing reference spectra. Due to the ‎needless stoping of machine for investigation of its various parts, use of this &l MoreThe base of diagnosing the possible defects of a machine is comparing the frequency ‎spectra of the vibrations at different points with the existing reference spectra. Due to the ‎needless stoping of machine for investigation of its various parts, use of this ‎troubleshooting method is affordable; Also, regarding to progress of possible ‎defectes, the machine can be rapaired in any required times. In this study , using ‎Neural Network (MLP and FNN), firstly common defects in rotating machines were created ‎separately, then the produced vibrational frequency were measured by ADASH 4400 ‎analyzer. Introducing four vibrational characteristics including angular misalignment, ‎clearance, failure and unbalance of bearing as input data of artificial neural network ,the ‎results were compared to the reference frequency signals. The results show that neural ‎networks MLP and FNN increase the defects detection ability by 73% and 78%, ‎respectively. So, FNN method is proposed for useful life prediction and detection of rotating ‎parts.‎ Manuscript profile -
Open Access Article
175 - مدلسازی لوله های انتقال گاز با شبکه های عصبی مصنوعی به منظور تشخیص عیوب آنها
علی جودکی مرتضی محمدظاهری احسان جمشیدی -
Open Access Article
176 - Fault Detection of Cylinder- head automotive Using Mechanical Waves and Intelligent Techniques
Mojtaba Doustmohammadi Morteza Mohammadzaheri Ehsan JamshidiSafety of machinery parts, it is important for users and defects as one of the most important parts of threats to the health of the pieces have always been considered. In this paper, a new method based on the use of artificial neural networkDetection Pridecylinder head MoreSafety of machinery parts, it is important for users and defects as one of the most important parts of threats to the health of the pieces have always been considered. In this paper, a new method based on the use of artificial neural networkDetection Pridecylinder head into mechanical waves is presented. Then, for verification and validation of the finite element model , modal test piece , and after confirming the fault simulation is performed on a finite element model. Force simulation model (FEM) acceleration signal - the healthy and faulty models calculated for each fault. And with the signing of mechanical defects ( acceleration difference between healthy and faulty models ) and to train a multilayer perceptron neural network (MLP) , the difference between the acceleration signal to the associated fault location. Detectionautomotive cylinder head (estimated fault location) has been. The results of the neural network capabilities designed to estimate the fault location on a good show. Manuscript profile -
Open Access Article
177 - آشکارسازی تغییرات کاربری اراضی و عوامل مؤثر بر آن با استفاده از شبکه عصبی مصنوعی(مورد مطالعه: شهرستان تالش)
شهرام امیرانتخابی فرهاد جوان حسن حسنی مقدم -
Open Access Article
178 - پهنه بندی مناطق در معرض پیشروی سطح آب دریا در اثر تغییر اقلیم (مطالعه موردی: بندر شهید رجایی)
حمید گوهرنژاد -
Open Access Article
179 - Identifying and reviewing the process of vegetation usage changes using time-based neural network and CA models using GIS and RS techniques (Case Study: Minoodasht County Golestan Province)
صادق شکوری سید مسعود موسوی حسنی مهسا پورعطاکش آناهیتا قربانی سمیرا ارنکMonitoring land use change is important in many planning and urban management activities. Due to human activities and natural phenomena, the face of the earth always changes.Therefore, for optimal management of natural areas, awareness of the land use change ratio is co MoreMonitoring land use change is important in many planning and urban management activities. Due to human activities and natural phenomena, the face of the earth always changes.Therefore, for optimal management of natural areas, awareness of the land use change ratio is considered necessary. The purpose of this study was to evaluate and reveal land use changes, especially the use of vegetation cover in the Auchan region, from the functions of Minoodasht city of Golestan province in a 30-year time span using remote sensing and spatial information systems and MATLAB, ARCGIS and ENVI software.For this purpose, Landsat satellite ETM sensor was used from 1987, 1993, 1998, 2000, 2003, 2008, 2013, 2015, and 2017, and after making necessary corrections in the preprocessing stage, to monitor vegetation time changes, the index Vegetation cover (NDVI) was calculated in MATLAB software for each 9 time intervals.Then, by using the calculated images of the first 7 years and the model of the neural network (time series), the images of the eighth and ninth year were predicted and obtained, and then calculating the RMSE error between the output images of the model with the actualImages, the validation model it turned out the results show that the model with an average RMSE of about 0.13 was very good for the NDVI.The CA model was used to predict vegetation changes. The results show that the vegetation cover in the last two years, 2015 and 2017, has been upgraded by the neural network model and the study area has become greener Manuscript profile -
Open Access Article
180 - Determining the amount of total organic carbon using satellite imagery and artificial neural network model (Case study area: Mountain Mish, Gachsaran city)
kamran mojardThe amount of Total Organic Carbon (TOC) is one of the most important parameter in evaluating hydrocarbon source rocks. Therefore, it is necessary to estimate of source rock by a method. There are several methods for estimating the of source rocks. One of the earliest m MoreThe amount of Total Organic Carbon (TOC) is one of the most important parameter in evaluating hydrocarbon source rocks. Therefore, it is necessary to estimate of source rock by a method. There are several methods for estimating the of source rocks. One of the earliest methods is the traditional method, which is very expensive and time consuming, so researchers are looking for more efficient methods. Given the high potential of remote sensing and spectral VIR spectra, the VNIR makes it possible to estimate the characteristics of the origin rock at a lower cost basis. Different quantitative and qualitative methods are used to establish a relationship between the concentration of rock elements and the spectrum obtained from remote sensing data. In this study, we tried to estimate the total organic carbon content of the total origin rock using the OLI Landsat 8 image sensor and using the MLP artificial neural network model. For this purpose, the band of 5 spectral ranges (0.845-0.8585) with Pearson correlation coefficient of 0.62 was chosen for the neural network. An artificial neural network with neurons in the secret layer with R2 = 0.79 and RMSE = 0.0008 were selected to generate a total organic carbon map. Manuscript profile -
Open Access Article
181 - Detection of land use changes using satellite imagery during the period 1984-2019 (Case study of Kamyaran city)
saman javaheri Ali asghar TorahiLand use change due to human activities is one of the important issues in regional and development planning. Given the advantages and capabilities of satellite data, this technology can be of great help in identifying and detecting these changes. The purpose of this stu MoreLand use change due to human activities is one of the important issues in regional and development planning. Given the advantages and capabilities of satellite data, this technology can be of great help in identifying and detecting these changes. The purpose of this study is to detect land use changes in Kamyaran city using satellite images over a period of 35 years. In this study, data from 1984 TM sensor, 2000 ETM + sensor and 2019 Landsat OLI sensor were used. Initially, preliminary preprocessions including radiometric, atmospheric and geometric corrections were performed on the raw data. Land control points were used for training, accreditation and to prepare land use map. Land use class was determined by field survey and using Google Earth images in 9 land use classes of agricultural lands, forests, gardens, rich and wooded pastures, medium rangelands, residential areas, water area, barren lands and rock outcrops. Next, the neural network method was used to monitor the images in ENVI 5.3 software. The evaluation results showed that the overall accuracy and kappa coefficient of OLI classified images are 94.3 and 0.92%, ETM + 92.6 and 0.91% and TM 90.3 and 0.87%, respectively. The results showed that forest lands and rich and wooded pastures decreased significantly during three time periods, which decreased by 11.64 and 19.12 percent, respectively. So that rich and wooded pastures have an increasing trend until 2000 and in the next period until 2019 has a decreasing trend. Residential lands, water areas and gardens increased by 2.27%, 0.57% and 3.98%, respectively. Due to the growing trend of population and urbanization, the results of this study provide the necessary information to make basic decisions in the development of management policies for planners and regional managers for the sustainability and evaluation of natural resources. Manuscript profile -
Open Access Article
182 - Quantitative studies in the management of the banking industry in order to increase customer satisfaction and profitability (case study: Bank Mellat)
Mohammad Moradi Mohammad Sadegh Horri Iraj NooriIn order to provide all kinds of facilities to their customers, credit institutions need to carry out complete surveys in order to know the applicants from qualitative and quantitative aspects, in order to fully evaluate the ability to repay and calculate the probabilit MoreIn order to provide all kinds of facilities to their customers, credit institutions need to carry out complete surveys in order to know the applicants from qualitative and quantitative aspects, in order to fully evaluate the ability to repay and calculate the probability of non-repayment of facilities and services. Financially, these surveys are generally called validation. The purpose of this research is to rank the groups of customers and determine the best parts of them so that the brokerage company can perform credit allocation in a mechanized way. For this purpose, after the initial pre-processing of the data, they are processed in the form of RFM 1 model. Then, using the SOM 2 neural network as one of the clustering algorithms, the customers will be divided into 10 clusters. In the following, using the proposed model, the clusters are ranked. The best clusters are identified and the operation of granting facilities is done for the members of these clusters. Finally, three clusters 5, 1 and 7 were determined as the best clusters, which are the target customers. The coefficient of facilities granted to these top three clusters is 0.271, 0.173 and 0.556 respectively. Manuscript profile -
Open Access Article
183 - پیش بینی اظهار نظر حسابداران رسمی با استفاده از شبکه های عصبی و رگرسیون لجستیک
بهرام همتی هاشم نیکومرام فریدون رهنمای رودپشتی رضا فرضی پور صائین -
Open Access Article
184 - تحلیل تغییرات کاربریهای اراضی نواحی حاشیه زاینده رود با مدلسازی در سامانه اطلاعات جغرافیایی (بازه چم خلیفه تا سامان شهرکرد)
علی اکبر جمالی سید علی المدرسی احسان ایزدی -
Open Access Article
185 - Identifying the Dimensions and Components of the Implementation of Information and Communication Technology Development Policies
Nazila Mohammadi Gholamreza Memarzadeh Tehran Sedigheh Tootian EsfahaniThe purpose of this research is to identify the dimensions and components of the implementation of information and communication technology policies with focus on the sixth development plan and to provide a model of the factors affecting the implementation with the help MoreThe purpose of this research is to identify the dimensions and components of the implementation of information and communication technology policies with focus on the sixth development plan and to provide a model of the factors affecting the implementation with the help of neural network modeling and based on Giddens constructive theory. This research is a survey and based on the goal is an applied type. Data collection is based on the library and field method with the questionnaire tool. In order to extract the effective factors, an expert panel including experts in the field of communication has been formed, and the statistical population of the research in the model test section is the information and communication technology experts of Iran Telecommunication Company (810 people), of which 260 people were randomly selected as samples based on Cochran's formula. MATLAB software was used for data analysis. According to the findings, the best combination for development is when all input variables are considered at the same time, and the worst case is when the infrastructure development variable is ignored, and the most important based on network sensitivity analysis is related to infrastructure development and the least related to content provision. Manuscript profile -
Open Access Article
186 - Providing a Model for Predicting the Success of Investment Projects in Free and Special Economic Zones, Using the Multi-Layer Neural Network Technique
morteza shokrzadeh kamaleddin rahmani farzin modares khiyabani majid bagherzadehTo analyze the data of this research, descriptive statistics and inferential statistics were used and experts selection software, MATLAB, SPSS and PLS software were employed Using theoretical foundations and libraries, six effective factors and variables predicting the MoreTo analyze the data of this research, descriptive statistics and inferential statistics were used and experts selection software, MATLAB, SPSS and PLS software were employed Using theoretical foundations and libraries, six effective factors and variables predicting the success or failure of Investment projects in the free and special economic zones of the country were identified.After describing the variables and testing the normality,using the PLS software, a confirmatory factor analysis of the variables was carried out, in which all of the factors had a good confirmatory factor analysis and all the questions were approved Then, using linear regression and ANOVA, the effect of each of the factors on the success or failure of investment projects was investigated, and the results of this test showed confirmation of the impact of each of the factors, and then the results of the hierarchical analysis indicated this was the first rank of product and service, followed by the second-rank ,that is geographical considerations, and the characteristics of the investor's psychology, the third rank, the product market characteristics, the fourth rank, the investor's ability to rank fifth, and financial considerations ,also, earned the last rank.Considering this prioritization, the neural network used in this research contained data from 6variables as an input variable, with two intermediate layers with 30 nodes in the first layer, and three nodes in the second layer, which had one outlet.The results indicated that the neural network model had the power to predict the success of the investment projects. Manuscript profile -
Open Access Article
187 - Using different learning algorithms in the stock price prediction by using neural networks
Reza Kiyani Mavi Kamran Sayadi NikStock price prediction is a very important financial topic, and is considered a challenging task and worthy of the considerableattention received from both researchers and practitioners. Stock price series have properties of high volatility, complexity,dynamics and turb MoreStock price prediction is a very important financial topic, and is considered a challenging task and worthy of the considerableattention received from both researchers and practitioners. Stock price series have properties of high volatility, complexity,dynamics and turbulence, thus the implicit relationship between the stock price and predictors is quite dynamic. Hence, it isdifficult to tackle the stock price prediction problems effectively by using only single soft computing technique.In this research, in the first step, the possibility of predicting stock price of National Iranian Copper Industries Company wasstudied. Then, for predicting of stock price after one day neural ¬network of MLP by learning algorithm of Levenberg-Marquardt were used. Then optimize structure of neural network was trained with the standard BP algorithm, the learningrate is 3/0 has the best performance. And for this learning rate, sensitive of standard BP algorithm was calculated to minimizelocal. At the end, standard BP algorithm with momentum is used. The results showed that predicting by standards BPalgorithm with momentum is better than the standard BP algorithm. Manuscript profile -
Open Access Article
188 - Evaluate the performance of the organization using two integrated approaches the DEA-BSC and ANN-DEA
Reza Ehtesham Rasi Eisa NajiConsidering the importance of efficiency in advancing communities and the place that today's organizations have embraced, the use of performance evaluation has become an inevitable necessity. Therefore, we are trying to investigate the effectiveness of organizations by MoreConsidering the importance of efficiency in advancing communities and the place that today's organizations have embraced, the use of performance evaluation has become an inevitable necessity. Therefore, we are trying to investigate the effectiveness of organizations by combining the two DEA-BSC systems and neural networks in this study. In this regard, first, the performance indicators were determined by using the Balanced Scorecard technique in four dimensions of customer, internal processes, learning and growth and financial, then using the data envelopment analysis method and non-radial SBM model and GAMS software, the organization's efficiency we calculated. In the next step, with the help of the indicators derived from the DEA-BSC method and using the combination of two systems of data envelopment analysis and neural network and using the MATLAB software, the efficiency was re-calculated. In the last step, we compared the two methods of DEA-BSC and ANN-DEA. The results of the comparison of the two methods indicated the compatibility of the two models in discussing the efficiency and superiority of the ANN-DEA method in terms of the short response time and the determination of the efficiency and the possibility of using Its trained algorithm to measure the performance of organizational units in the future. Manuscript profile -
Open Access Article
189 - Assessing Credit Risk in the Banking System Using Data Mining Techniques
Nima Hamta Mohammad Ehsanifar Bahareh MohammadiA credit risk is the risk of default on a debt that may arise from a borrower failing to make required payments. The objective of this paper is recognition of the factors that effect on credit risk and presenting a model for prediction of credit risk and legal customer MoreA credit risk is the risk of default on a debt that may arise from a borrower failing to make required payments. The objective of this paper is recognition of the factors that effect on credit risk and presenting a model for prediction of credit risk and legal customer credit ranking that are applicant of Sepah bank facilities in Dezfool city and the method of Clustering, Neural Network and Supporter Vector Machine has been used in the current study. Accordingly necessary investigations have been done on financial and nonfinancial data by means of a simple random sample of 200 legal customers that were applicant of bank facilities. In the this paper, 27 descriptive variable that include financial and nonfinancial variables were investigated and finally available variables 8 effective variables on credit risk were selected by means of bank experts judges that were separated by data collection Clustering method in to some groups (Clusters) in the someway that data in one Cluster were considering other points in other Clusters had more similarity. Also selected variables with 3 layers perceptron Neural Network input vector entered the model and finally by means of Support Vector Machine was presented in order to bank legal customers’ financial operation prediction. The obtained results of Neural Network model and Supporter Machine indicate that Neural Network model has mire efficiency in legal customers’ credit risk prediction and credit ranking. Manuscript profile -
Open Access Article
190 - Improving the Efficiency of Forecasting Productivity, Using a Taguchi Experiment Design Approach (Case Study: Food Industries in Iran)
Seyed mahmon Zanjirchi Mehdi Hatamimanesh Hamedreza Kadkhodazadeh Seyedali MohammadbanifatmiProductivity forecasting is a key factor in strategy planning in an organization. Artificial neural networks method is one of the productivity estimating methods whose users must have enough experience and skill because of its adjustable parameters. Trial and Error is m MoreProductivity forecasting is a key factor in strategy planning in an organization. Artificial neural networks method is one of the productivity estimating methods whose users must have enough experience and skill because of its adjustable parameters. Trial and Error is mostly used to find the proper levels of these parameters. This article presents a seven step pattern for selecting proper adjustable parameters for neural network, using Taguchi experiment design method to improve the efficiency of productivity forecasting. As a result, the optimum parameters levels that lead to the most desirable forecasting in neural network are as follows: the number of hidden layers: 2 layers, the number of neurons in each hidden layer: 7 neurons, learning rate: 0.9 and the number of neural network inputs: productivity indicators with more than 0.85 degree of correlation. Among the above mentioned factors, the number of hidden layers with 71.18% of contribution rate in experiment results is the most important factor in neural network design to forecast the productivity of Iranian food industry. Finally, the overall results of the study showed that using this pattern provides the possibility of choosing competitive strategies besides decreasing forecasting time and cost. Moreover, this pattern helps decision makers with the extent of the consideration that must be put into each adjustable parameter by determining the contribution rate of each parameter in the experiment results. Manuscript profile -
Open Access Article
191 - The Prediction of the Success of New Product Development Using the Combination of Factor Analysis and Artificial Neural Network
Gholamraza Soltani Fesaghandis Alireza Pooya Mostafa Kazemi Zahra Naji AzimiThe success of new products is as the greatest element for the company's success and even their survival. However, numerous studies show that the failure rate of new product development as a fundamental problem in many companies has been raised. Many companies are tryin MoreThe success of new products is as the greatest element for the company's success and even their survival. However, numerous studies show that the failure rate of new product development as a fundamental problem in many companies has been raised. Many companies are trying to predict the success of the development of new product before its development. The aim of this study was to predict the success of new product development using a combination of factor analysis and artificial neural network in the food and beverage industry. This research based on purpose is practical and based on method is descriptive. The population of this research is the food and beverage manufacturers in the province of East Azerbaijan in 1394. In order to collect data two questionnaires have been distributed among the population after the validity and reliability assessments. In order to analyze the data in this study, factor analysis and artificial neural network is used. Analysis of the data revealed the presence of six main structures as factors in the success of new product development: conceptualization of new product, market orientation, design orientation, technology orientation, use of sources, and management of new product development as the effective elements in the development of the success of new product. Furthermore, the prediction of the success of new product’s development using neural networks shows that the designed network was able to predict the success of the development of new product correctly. Manuscript profile -
Open Access Article
192 - Study of Relationship between North Atlantic Teleconnections and Air Temperature in Caspian Sea Basin.
Iraj Heidari Amir Gandomkar Mohsen Bagheriپژوهش حاضر با هدف بررسی تأثیر الگوهای پیوند از دور بر میانگین دمای حوضه آبریز دریای مازندران صورت پذیرفته است. در این راستا از آمار میانگین دمای 97 ایستگاه همدیدی و اقلیم شناسی و همچنین داده های 33 الگوی پیوند از دور طی دوره 2014- 1970 در مقیاس ماهانه و سالانه استفاده ش Moreپژوهش حاضر با هدف بررسی تأثیر الگوهای پیوند از دور بر میانگین دمای حوضه آبریز دریای مازندران صورت پذیرفته است. در این راستا از آمار میانگین دمای 97 ایستگاه همدیدی و اقلیم شناسی و همچنین داده های 33 الگوی پیوند از دور طی دوره 2014- 1970 در مقیاس ماهانه و سالانه استفاده شد. پس از تایید نرمال بودن داده ها توسط آزمون ران تست داده های ایستگاهی به روش کریجینگ به داده های نقطه ای با ابعاد 7/19 × 7/19 کیلومتر تبدیل شدند. به منظور ارتباط سنجی متغیرها از آزمون های پیرسون، رگرسیون خطی و مدل شبکه عصبی استفاده شد. نتایج حاصل نشان داد الگوهای واقع در اقیانوس هند و آرام و الگوهای نیمکره جنوبی، رابطه معنی دار چندانی با نوسان های دما در حوضه مازندران ندارد. در مقابل، الگوهای پیوند از دور مستقر در اقیانوس اطلس و قطب شمال ارتباط زیادی با نوسان های دما در حوضه دارد. شایان ذکر است از بین الگوهای پیوند از دور سه الگوی دریای شمال- کاسپین، نوسان اطلس شمالی و نوسان قطبی بیشترین رابطه را با نوسان های دمای ماهانه و سالانه در این حوضه دارند. Manuscript profile -
Open Access Article
193 - Comparing the Performance of Neural Networks and Multivariate Regression in the Estimation of Housing Prices (Case Study: Ahvaz City)
said amanpour ismail soleimanirad leila keshtkar Sadegh MokhtariHousing is a basic need in the community always. Housing market has been one of the most fluctuation sectors of the economy of Iran country over the past years. Since the housing sector changes have a great impact on other sectors of the economy, so one of the significa MoreHousing is a basic need in the community always. Housing market has been one of the most fluctuation sectors of the economy of Iran country over the past years. Since the housing sector changes have a great impact on other sectors of the economy, so one of the significant needs of housing is predicting the price of this good. In this context, in this study by using of multi-layer perceptron neural network, presented a model to predict housing price in the city of Ahvaz and the results compared with the multivariate regression model. This study is a practical–developmental and its method is analytical- descriptive. To do this, 233 samples of statistical unit in 1392 were analyzed on the basis of 16 relevant variables. The results show that multi-layer neural network with 91 percent accuracy have been more accurate compared with the multivariate regression in the predicting housing prices. In addition to evaluating the performance of models coefficients R^2 and RMSE were used. Coefficient of determination (R^2) by using multivariate regression is .789 and its value for neural network is .918. The result of the regression model indicates weaker performance of this model compared to artificial neural network approach. Manuscript profile -
Open Access Article
194 - Explaining the Optimal Model of Smart with Emphasis on Improving the Business Structure with the Neural Network Model in Yazd
Amir Poorrajayi Seyed Ali Almodaresi Mohammad Hosiyn Sarayi Ahmad EsteghlalSmart growth as a comprehensive strategy to counter the spread of sporadic and low-density areas around cities was discussed. The purpose of this study was to prioritize different areas of the research area and identify areas with intelligent potential in order to apply MoreSmart growth as a comprehensive strategy to counter the spread of sporadic and low-density areas around cities was discussed. The purpose of this study was to prioritize different areas of the research area and identify areas with intelligent potential in order to apply management strategies. To achieve this goal, 8 main variables of population density, per capita business use, use of counter offices, type of Internet and road network coverage, per capita residential, parking, bank were used.151 GPS points were taken to model and evaluate the validity of the model. Raster data layers to enter the model and neural network modeling was modeled Yazd smart prone areas. The research results showed regression above 70% in the training and testing phase. The area under the curve AUC validation model actually reflects the ability of the model to predict the dependent variable was 0.9769, which was very convenient. The parameters MSE, RMSE, R_Correlation_Test and R2 were also obtained as 0.0389, 0.1972, 0.8517 and 0.8912, respectively. Finally, the weight of the indicators and the dependent variables of the research were predicted. The results of the final map of area zoning capabilities in smart too low to 4013.78 showed that telling this story for intelligent infrastructure must be strengthened in this area. Areas with very high potential with an area of 687.31 have also been notable. By examining the modeling map and GPS points of Areas prone to smartening, a high compliance in the modeling was performed and the field points were observed. Manuscript profile -
Open Access Article
195 - Study and Evaluation of Temperature in Aleshtar City based on Artificial Neural Network Model
Mahnaz Hassanvand Reza Borna Manijeh Zohoorian Pordel Alireza ShakibaTemperature assessment and forecasting is one of the most practical estimates of climatic elements. Today, the agricultural and industrial sectors are highly dependent on the temperature conditions. Temperature is one of the most important climatic meters that is one of MoreTemperature assessment and forecasting is one of the most practical estimates of climatic elements. Today, the agricultural and industrial sectors are highly dependent on the temperature conditions. Temperature is one of the most important climatic meters that is one of the main factors in the climate identity of each region. The purpose of this study is to make a model for predicting the average monthly seasonal temperature of selected stations in Lorestan province, including Al-Shatrami region. Identification and detection of vulnerabilities in the infrastructure of Aleshtar districts in the conditions of climate change. And due to the inadequacy of the 30-year time series of Al-Ashtarl, neighboring cities such as Khorramabad-Aleshtar-Borujerd synoptic stations have been used, because the artificial neural network method has a great ability to simulate and predict atmospheric elements. And the weather, especially the temperature. To model and predict the seasonal monthly temperature, the r programming tool software of the fOre gast package has been used. Two tests of estimator trend analysis have been used. The 30-year time series trend of these elements was examined during the basic statistical period (1989-2019). The climate cycle was reported and extracted under two scenarios: NNAR and forEgast. The artificial neural network is one of the most powerful models capable of receiving and displaying complex Data input and output is one of the most widely used neural network (NNA) models to determine the best network inputs. Manuscript profile -
Open Access Article
196 - On the Current Changes of Votes Between 1989 and 2019 Changes and Predicting the Changes Using Remote Sensing and CA-Markov and LCM Models
Behroz Sadayn Mohammad Ebrahim AfifiLand use change has acted as a pivotal factor in environmental change and has become a global threat. Reviewing these changes through satellite imagery and predicting and evaluating their potential through modeling can help environmental planners and natural resource ma MoreLand use change has acted as a pivotal factor in environmental change and has become a global threat. Reviewing these changes through satellite imagery and predicting and evaluating their potential through modeling can help environmental planners and natural resource managers to make informed decisions. The purpose of this study was to review, model, and predict land use changes in the 30-year period of 1993-2013 by the Markow-LCM chain model in Kangan and Assaluyeh. For this purpose, land use maps were prepared using ETM +, TM and OLI satellite imagery in three periods of 1993, 2003, and 2013. Then verifying the maps and detecting the changes. Using the classification of the neural network and applying the Land Change Modeler (LCM Markov model) and the Land Use Change Modeling Approach have been implemented. The results of detection of changes in the first period with a kappa coefficient of 97% and the second period of 1993-2003 with a kappa coefficient of 94% indicate that the largest changes in the area in the water area and the largest decrease in the area in the vegetation area occurred. In order to calibrate the Markov chain model, the 2013 map was predicted and the error mapping matrix of the 2013 map reference model and mapping utilization yielded a copper coefficient of 93%. The results of modeling the transfer force using the artificial neural network in most of the sub-models The high accuracy was 60-93%. Manuscript profile -
Open Access Article
197 - Environmental sustainability assessment with emphasis on drought and water resources using multi-criteria artificial neural network technology (case study of Babak city)
محمد ابراهیم عفیفی ahmad mangeli meydook ali vakhsooriIn this study, first, by studying the research, criteria and sub-criteria were identified that are effective in terms of environmental sustainability. After the Delphi stages, the criteria of environmental resources and services, environmental health and energy were sel MoreIn this study, first, by studying the research, criteria and sub-criteria were identified that are effective in terms of environmental sustainability. After the Delphi stages, the criteria of environmental resources and services, environmental health and energy were selected as the most important criteria for assessing environmental sustainability in Babak, then using the neural network model to analyze and evaluate the environmental sustainability of Babak. In this study, drought in Babak city was analyzed with a SPI index of drought during a statistical period of 32 years 1361-1392. This index is specifically for time series six; Twelve and forty-eight months were calculated. The city of Babak has been facing drought during the statistical period of thirty-two years, especially the last seven years, and on an annual scale of six months, most of its droughts are mild to moderate droughts. But in the long-term Myas 48 months, 75% of the droughts were severe and very severe, which shows a high relationship with the quantitative and qualitative decline of groundwater in this area. Manuscript profile -
Open Access Article
198 - Evaluation of environmental sustainability in urban areas with the approach of fine dust problem using multi-criteria decision making technique of linear allocation and artificial neural network (case study: Ahvaz)
yahya abdolkarim neysi Mohammad Ebrahim Afifi marzeyeeh mogholiEnvironmental sustainability assessment is one of the most important tools in the process of sustainable development planning and therefore attention to it in policy-making and planning is inevitable. The purpose of this paper is to evaluate the environmental sustainabi MoreEnvironmental sustainability assessment is one of the most important tools in the process of sustainable development planning and therefore attention to it in policy-making and planning is inevitable. The purpose of this paper is to evaluate the environmental sustainability in urban areas with the approach of the dust problem using multi-criteria decision-making technique of linear allocation and artificial neural network in Ahvaz metropolis. In this research, first, by studying the theoretical foundations of the research, the criteria that are effective for environmental sustainability were identified. Then, the documentary and library method and referring to different departments and organizations were used to collect information in the theoretical part. However, the main information of the research was collected using field studies (completing questionnaires, observations and field studies). The research process was such that a questionnaire was designed to implement the methodology through the criteria obtained from the study of the problem literature and distributed among the people in 8 regions by stratified random sampling method. The sample size of the total areas was determined based on the unlimited Cochran's formula in the form of population classes of the areas. Keywords: Environmental stability, particulate matter, linear allocation, artificial neural network, Ahvaz Manuscript profile -
Open Access Article
199 - The zoning of slop instabilities on the mountain roads by Artificial Neural Network (MLP)(case study: dare diz strait)
Shahram Roostaei Fatemeh khodaeiDre Diz straitis one of the most risky straits of east Azarbayjian province from occurrence of slop instabilities. Given، the coincidence of thisstrait on the only connection pass between MARAND_JOLFA towns، the best way to care and keeping the security of the road، is MoreDre Diz straitis one of the most risky straits of east Azarbayjian province from occurrence of slop instabilities. Given، the coincidence of thisstrait on the only connection pass between MARAND_JOLFA towns، the best way to care and keeping the security of the road، is identification of risky areas or zoning the risk of slop instability in this strait. For this ، the main data was collected then the needful layers were provided on the GIS software environment ،later the zoning map of slop instability for zoning slop instability risk obtained in IDRISI software and on the ANN method (MLP) with 1-10-15 instruction provided ،and the apt areas for occurrence of slop instability have been introduced in five different risk class :highest، high، medium، low ،very low.And according to the result slop and distance of faulthave the greater role on the happining of slop instabilities on the strait and therefor on the Insecurityof the road. Manuscript profile -
Open Access Article
200 - Evaluating the effects of climate change on Lar Basin Water Resources Using SWAT Model and comparing its results with Bayesian Networks and Hybrid Intelligent Models
Mahsa Solimani Puor Amirpouya SarrafIran's location on the arid and semi-arid belt of the world, as well as the mismanagement of water resources, has created a warning situation of water shortage in many parts of the country. The present research evaluates the effects of climate change on temperature, rai MoreIran's location on the arid and semi-arid belt of the world, as well as the mismanagement of water resources, has created a warning situation of water shortage in many parts of the country. The present research evaluates the effects of climate change on temperature, rainfall and runoff in future periods with the help of LARS-WG statistical model and SWAT hydrological conceptual model for Lar Basin. To estimate the flow rate of the river, the performance of Bayesian network and the combined wavelet-neural network model are also examined. After entering the rainfall and temperature information of the region, runoff was simulated for two hydrometric stations of Gozeldareh and Plour and the outflow runoff of Plour station between 1979 to 2018 was calibrated and validated as a control point. In order to evaluate the efficiency of the models, the criteria of Nash-Sutcliffe and explanation coefficient are used. According to climate models, the highest temperature increase in the final period and under the RCP8.5 climate scenario shows about 10% increase in temperature in spring and winter. Finally, among these models, the physical model with an average annual prediction of 6.04 cubic meters per second according to the observation period, showed a decrease in runoff. Manuscript profile -
Open Access Article
201 - Change Time Study Pricipitain of Hamadan Province Using Statistic’s and Neural Network Methods
Zohreh Maryanaji Hamed AbbasiClimate change is one of the common problems in today’s societies and considerate as threat for earth. Increasing of earth temperature is cased to extensive changes in earth climates and also it laded to same variations of spatial and temporal of precipitation so MoreClimate change is one of the common problems in today’s societies and considerate as threat for earth. Increasing of earth temperature is cased to extensive changes in earth climates and also it laded to same variations of spatial and temporal of precipitation so that these variations cased to a lot of damages especially in last decade. This study to identify of variations and the trend of season and duration of precipitation in different times intervals. Then it is predicted same changes in the future by the method of artificial neural network. In this study we have us of the data from two synoptic stations Hamadan and Nojeh .The statistic’s years in this study; it seems the rainfall season in the central part of Hamadan province in started later and ended later in last decade. In other words the starter of rainfall season in Hamadan which occurred in the fall season, at present tend to ward the winter season and the fall season is more dryer than before ages. This shows that the rainfall season is interchanged in this district. in using of the method of artificial neural network we should consider to two main points in the predicting of precipitation the first one, the low attention of this method in the long –term predicting of precipitation and the second one, the exaggerate in the minimum and maximum amount of precipitation in different seasons of year. Manuscript profile -
Open Access Article
202 - تحلیل ارتباط الگوهای پیوند از دور با خشکسالی حوضه قرهقوم با استفاده از مدل شبکه عصبی
مونا فلاحزاده پرویز رضایی سعید اسلامیان علیرضا عباسیدر این پژوهش نقش الگوهای پیوند از دور در رخداد خشکسالیهای حوضه قرهقوم مورد بررسی قرار گرفته است. در این راستا دادههای بارش 30 ایستگاه بارانسنجی و سینوپتیک و نیز دادههای مربوط به 32 نمایه عددی پیوند از دور از سایت نوآ طی دوره آماری 1987-2013 اخذ گردید. در ابتدا داد Moreدر این پژوهش نقش الگوهای پیوند از دور در رخداد خشکسالیهای حوضه قرهقوم مورد بررسی قرار گرفته است. در این راستا دادههای بارش 30 ایستگاه بارانسنجی و سینوپتیک و نیز دادههای مربوط به 32 نمایه عددی پیوند از دور از سایت نوآ طی دوره آماری 1987-2013 اخذ گردید. در ابتدا دادههای شاخص بارش استاندارده شده با روش تحلیل عاملی طبقهبندی، سپس رابطه میانگین شاخص خشکسالی هر پهنه با تک تک الگوهای پیوند از دور ارزیابی شد. همچنین مقادیر شاخص خشکسالی با شاخصهای از دور به روش شبکه عصبی مصنوعی شبیهسازی گردید. نتایج نشان داد 5 شاخص پیوند از دور نوسان دههای اقیانوس آرام، نینو4، چند متغیره انسو، دو قطبی اقیانوس هند و نوسان مادن جولیان در منطقه 1 اقیانوس آرام با پهنه اول (عامل اول) در مقیاس زمانی 6 ماهه در ارتباط بوده و بهترین نتایج را با کمترین خطا و بیشترین ضریب همبستگی ارائه دادهاند. Manuscript profile -
Open Access Article
203 - پیشبینی بارش فصلی با حداقل متغیرهای اقلیمی مطالعه موردی: ایستگاه کرمان
Fateme Bayatani غلام عباس فلاح قالهری غلام عباس فلاح قالهری الهام فهیمی نژاد الهام فهیمی نژاد پیش­بینی بارش و برآورد نزولات جوی، به عنوان یکی از مهمترین پارامترهای اقلیمی در حوزه مدیریت منابع آبی، از اهمیت ویژه­ای برخوردار است. بنابراین در این مقاله، امکان کاربرد شبکه عصبی در برآورد بارش با حداقل پارامترهای اقلیمی مورد بررسی قرار گرفت. به این منظو More پیش­بینی بارش و برآورد نزولات جوی، به عنوان یکی از مهمترین پارامترهای اقلیمی در حوزه مدیریت منابع آبی، از اهمیت ویژه­ای برخوردار است. بنابراین در این مقاله، امکان کاربرد شبکه عصبی در برآورد بارش با حداقل پارامترهای اقلیمی مورد بررسی قرار گرفت. به این منظور از شبکه عصبی پرسپترون چند لایه با قانون پس انتشار خطا و الگوریتم سیگموئید همراه با داده های میانگین رطوبت نسبی(meanHR)، کمینه رطوبت نسبی (minHR)، بیشینه رطوبت نسبی (maxHR)، میانگین دما (meanT)، کمینه دما (minT)، بیشینه دما (maxT)، میانگین فشار (meanP)، کمینه فشار (minP) و بیشینه فشار (maxP) ماه اکتبر ایستگاه هواشناسی سینوپتیک کرمان، طی دوره آماری 2014-1969 به عنوان ورودی مدل استفاده گردید. نتایج نشان داد در صورت کمبود پارامترهای اقلیمی، تنها با اندازه گیری minT و meanT میتوان با خطایی معادل 8/9 میلیمتر، برآورد مناسبی از بارش با استفاده از شبکه­های عصبی مصنوعی در منطقه مورد مطالعه به دست آورد. Manuscript profile -
Open Access Article
204 - پیش بینی وقوع بارش روزانه با استفاده از داده های هواشناسی روزهای قبل (مطالعه موردی: شهر اصفهان)
Ghorban Mahtabi فرشید تاران سعید مظفریهدف از این تحقیق، پیش­بینی وقوع بارش روزانه شهر اصفهان با استفاده از داده­های هواشناسی 1 تا 7 روز قبل می­باشد. برای این منظور،داده­های هواشناسی دوره 2009-2000 با استفاده از مدل­های هوشمند بردار پشتیبان، k-نزدیک­ترین همسایگی، شبکه عصبی مصنوعی و در Moreهدف از این تحقیق، پیش­بینی وقوع بارش روزانه شهر اصفهان با استفاده از داده­های هواشناسی 1 تا 7 روز قبل می­باشد. برای این منظور،داده­های هواشناسی دوره 2009-2000 با استفاده از مدل­های هوشمند بردار پشتیبان، k-نزدیک­ترین همسایگی، شبکه عصبی مصنوعی و درخت تصمیم بررسیگردید. نتایج نشان داد که در هر چهار روش، دقت پیش­بینی بهترین سناریوها با استفاده از داده­های 6 و 7 روز قبل، کمتر از 75 درصد بود، اما با استفاده از داده­های روزهای 1 تا 5 روز قبل، بارش روزانه با دقت بیش از 80 درصد پیش­بینی شد. عملکرد روش درخت تصمیم بهتر از سه روش دیگر بود و به علت ارائه درخت تصمیم­گیری، نتایج سناریوهای 1 تا 5 روز قبل این روش ارائه شد. نتایج سناریوها با استفاده از داده­های 1 تا 3 روز قبل نشان داد که رطوبت نسبی هوا مناسب­ترین پارامتر برای پیش­بینی وقوع بارش روزانه است، اما در شرایط استفاده از داده­های 4 و 5 روز قبل، دمای هوا مناسب­ترین پارامتر برای انجام پیش­بینی بود. در نهایت عملکرد بهترین سناریوها با استفاده از داده­های دوره 2016-2010 صحت­سنجی گردید. بهترین نتایج در بخش صحت­سنجی به ترتیب مربوط به سناریوی 1 روز قبل(با پارامتر حداقل رطوبت نسبی) و سناریوی 4 روز قبل(با پارامتر دمای حداکثر) بود. Manuscript profile -
Open Access Article
205 - Comparsion Of Experimental, Regression Models and Artificial Neure Network in Estimating Net Radiation (Rs) In Synoptic Station of Zahedan
Parisa kahkhamoghadam mohammad mahdi chariSolar radiation is one of the key inputs for most hydrological models in estimating reference evapotranspiration. Furthermore providing and making the measurement tools for this parameter is very costly. In this research, ridation (Rs ) of zahedan meteological station i MoreSolar radiation is one of the key inputs for most hydrological models in estimating reference evapotranspiration. Furthermore providing and making the measurement tools for this parameter is very costly. In this research, ridation (Rs ) of zahedan meteological station in 1385 to 1389 were used. Some non- linear models such as neure systemwith algorithm BFGS, and neure system with conjugate Gradient training algorithms, and locallinear regression through gamma test were developed. Then , these non- linear models and two expereimental model including Angstrom - Prescott and Glory Mac Kalut were assessed for predicting radiation. For predicting none- linear method, maximum temperature parameters, average speed of wind, surface radiation, and Sunshine were used. Result of comparing measured amounts with models with measured amount by parameter show that the neure system with BFGS algorithm has RMSE= 1.95 , MAE= 1.47 and R2=93% which are the best operation in these models. After that, neure system model with conjugate Gradient training algorithms and local regression model are in secand rank in which RMSE, MAE and R2 are 2.53 , 1.77 , 88% and 2.89 , 1.89 , 82% respectively. Angstrom and MAC colt method have RNSE = 4.38 , MAE=3.21 , R2=33% and RMSE= 4.46, MAE= 3.07, R2=50% respectivety. Manuscript profile -
Open Access Article
206 - land use change modeling using LCM module (Case study: NEKA region)
seyede massoomeh fathollahi roudbary Kamran Nasirahmadi mehrdad khanmohamadiLand use/cover changes, especially human urbanization Cause destruction of natural habitats and threaten biodiversity. Regularly, Land use/cover models are one the most important methods for evaluating this trend. The objective of this study is the investigation of land MoreLand use/cover changes, especially human urbanization Cause destruction of natural habitats and threaten biodiversity. Regularly, Land use/cover models are one the most important methods for evaluating this trend. The objective of this study is the investigation of land use/cover change and modeling in the Neka city using Land Change Modeler (LCM). Landsat TM (١٩٨8), ETM+ (٢٠٠2), and OLI (2016) data was used for land use/cover classification and change. In addition, transition potential modeling was conducted using an artificial neural network. In this method, 5 sub-models and 9 variables were used. Then calibration period (1988-2002) was used by Markov chain and hard prediction for extrapolating the 2016 land use/cover changes. Finally, land use/cover maps for 2002 and 2016 were used for land use/cover map extending prediction to the year 2030. The accuracy assessment of model was conducted by Error Matrix. The results of this study showed the annual rate of decline in the forest was 2297 Hectare during the period 1988-2016. The biggest changes were in the conversion of forest lands to agriculture. Modeling results using artificial neural network also showed acceptable accuracy (69%). The results of modeling for 2030 also showed that the area of the forest is decreasing, Agricultural lands and urban areas are increasing. Manuscript profile -
Open Access Article
207 - Investigating of Hydro-geochemical of Groundwater in Shiramin Plain using Artificial Neural Networks
Reza Norouzi Somayeh Emami Hamed ShiralizadhehGroundwater is one of the most important water resources in arid and semi-arid regions. Increased water consumption due to population growth, has a great impact on quality and quantity of water supply. The main aim of this study was to evaluate the accuracy of interpola MoreGroundwater is one of the most important water resources in arid and semi-arid regions. Increased water consumption due to population growth, has a great impact on quality and quantity of water supply. The main aim of this study was to evaluate the accuracy of interpolation methods for predicting the spatial distribution of some groundwater quality indices for Shiramin plain. In this study, hydro-geochemical quality of groundwater resources for drinking, agricultural and industry purposes was investigated with the Wilcox and Schoeller Standards in Shiramin plain East Azerbaijan province, Iran. Further in the research, the modeling of quality parameters (TDS), EC and SAR is done with using artificial neural network. According to Schuler and Wilcox groundwater quality index, water was moderately suitable and unsuitable for agriculture and drinking, respectively. The most of the samples were in C3-S1 category. The results are representative of the acceptable performance of ANNs to predict groundwater quality. . Manuscript profile -
Open Access Article
208 - Investigation of Some Ecological Characteristics of Artemisia sieberi Besser. and Estimation of its Density by Neural Networks in Roodab Section of Sabzevar
Alireza Ghasemi Arian fazel fazeli hossein rohaniAbstract Artemisia sieberi Besser. (Asteraceae) is a medicinal plant belongs to flora of Iran , which is used in treatment of parasitic and infectious diseases. In recent years, many habitats parts of this plant have been destroyed by human activities. The purpose of th MoreAbstract Artemisia sieberi Besser. (Asteraceae) is a medicinal plant belongs to flora of Iran , which is used in treatment of parasitic and infectious diseases. In recent years, many habitats parts of this plant have been destroyed by human activities. The purpose of this study was to evaluate the autecological characteristics and estimation of Artemisia density in Roodab region of Sabzevar in 2017. At first the A.sieberi habitats was determined on satellite images, then ecological information including topography, climate, soil, geology and phenology was collected. In the next step, Artemisia densities in degraded rangelands were investigated by using neural network model. For this purpose, 70 random soil samples consisting of independent variables (texture, EC, SAR, pH, N, P, K, cations, organic matter and lime percentage) as well as dependent variable (plant density) were used to construct the model. The results were showed that A.sieberi prefers altitudes of of 1400 to 1800 m, 0% to 12% slope, 160 to 200 mm rainfall, and loamy to calcareous loam soils with low salinity. plant vegetative activity begins in late March and seed ripening occurs early in December. The results of model prediction indicated that the lowest plant density with 0.11 / m2 was related to the plots that were 40 years under plowing stress and the highest density with 0.4 /m2 was related to the plots after two years of degradation. The regression model (R2) showed that 95% of independent variables were involved in determining plant density. Model prediction also indicated that the lowest density by 0.11 and 0.4 plant per m2 belonged to rangelands have been plowed more than 40 years and which lasted two years from their destruction, respectively. Regression model (R2) also showed that independent variables have a 95% effect on the determination of A.sieberi density. Manuscript profile -
Open Access Article
209 - Assessing the development level of rural areas of central district of Falavarjan County: The application of Artificial Neural Network
Hamid Rastegari Mehdi NooripoorPlanning of development programs either at macro or at micro levels, needs to obtain a proper understanding about the differences between rural areas in terms of their infrastructure, social, economic and environmental aspects; so that this understanding can contributed MorePlanning of development programs either at macro or at micro levels, needs to obtain a proper understanding about the differences between rural areas in terms of their infrastructure, social, economic and environmental aspects; so that this understanding can contributed to policymakers and decision-makers the ability to recognize abilities, differences and deprivations of rural areas in order to organize a balanced development in the supposed areas. Therefore, the aim of this research was to assess the level of development of rural areas of central district of Falavarjan County. A survey research method was used in this study. Research population included villages with 20 and more households (26 villages) that about 376 household of these villages were selected as research sample based on Cochran Sampling. A researcher constructed questionnaire and secondary data taken from the village ID in 2010 was used to collect data that its face validity procedure was verified by faculty members of the Rural Development Management Department of Yasouj University and its reliability was also verified calculating Cronbach's Alpha reliability (0.65 ≤∝≤ 0.95). In order to assess and prioritization rural development, 24 indices in four dimensions were used: social-cultural, economic, environmental and physical-infrastructure. In order to analyze the data, the raw data were normalized and then using Artificial-Neural Network (ANN) of multi-layer feed-forward back-propagation, weight of each rural development index was obtained, respectively. Data processing was done using MATLABR2015a and SPSS22 software. The results showed that Jojil, Jowlorestan and Zefreh ranked first to third respectively and Mehrenjan Atrak and Mehrenjan ranked last in terms of rural development. The overall results of this study showed that the rural development is in a relatively favorable situation in the studied villages. Manuscript profile -
Open Access Article
210 - پیشبینی سرمای دیررس بهاره با استفاده از شبکهی عصبی پرسپترون چند لایه (MLP) و تاثیر آن در حمل و نقل شهر خرمآباد
Saeid Taghavi Haniyeh Omidzadeh -
Open Access Article
211 - Land Use Mapping of Sabzevar using Maximum Likelihood and Artificial Multilayer Perceptron Neural Network
Elahe Akbari Majid Ebrahimi Abolghasem AmirAhmadiAmong the important factors in urban planning and management, particularly in line with the achievement of the sustainable development in the urban areas as well as regarding the optimal use of the land, is on-time access to the data of land cover conditions in these re MoreAmong the important factors in urban planning and management, particularly in line with the achievement of the sustainable development in the urban areas as well as regarding the optimal use of the land, is on-time access to the data of land cover conditions in these regions. The remote sensing data has a high potential for the preparation of the update urban land cover maps. In order to present on-time and digital satellite data, a variety of shapes and possibility of processing during land cover maps are of high significance. In order to use the satellite photos Landsat/ETM+ and two algorithm of supervised classification including the maximum likelihood and the artificial neural network, land cover maps were prepared. During classification, the neural network algorithm of a perceptron network with a hidden layer and 7 input neurons, nine middle neurons and 4 output neurons were used. The input neurons are the same in number as the bands of the Landsat photos and the number of output neurons are the same as land cover map classes. Eventually, land cover map of the region has been classified into four classes of residential areas, barren lands, plant coverage, and roads. In order to evaluate the correctness of the classification results, many photos have been taken using GPS. Using overall accuracy and Kappa Coefficient the precision evaluation results of these two methods indicate that perceptron neural network has an overall accuracy of 98/24 and Kappa Coefficient 97/03 compared to the algorithm of maximum likelihood with an overall accuracy of 94/23 and Kappa Coefficient 90 / 34 is of higher precision. The findings of this study also show that the classification method for multilayer perceptron neural network as compared with the maximum likelihood method is of higher separation and capability for preparing the land cover map in the urban regions. Manuscript profile -
Open Access Article
212 - Identifying Factors Affecting Non-curent Debts of Banks Using Neural Networks and Support Vector Machine Algorithm
sajjad kordmanjiri iman dadashi zahra Khoshnood hamid reza gholamnia roshanThe main purpose of this paper is to identify the factors influencing the creation and increase of non-current debts to make a more appropriate decision in granting facilities. For this purpose, to select effective variables, from the analysis algorithms of correlation MoreThe main purpose of this paper is to identify the factors influencing the creation and increase of non-current debts to make a more appropriate decision in granting facilities. For this purpose, to select effective variables, from the analysis algorithms of correlation and Lasso components; And to classify the samples, neural networks and support machine were used. In this study, a sample of 660 legal customers of Sepah Bank for the years 2006-2017 was selected and focused on the characteristic variables extracted from the facility contracts of these customers along with financial, non-financial, auditing and economic variables. The results showed that the Lasso algorithm focused on financial, economic and auditing variables, performed better than the neighboring component analysis algorithm, and based on this algorithm, 10 key variables affecting non-current debts were identified. Due to the better performance of support vector machines with radial cores, its use in modeling non-current debts is recommended. Manuscript profile -
Open Access Article
213 - Modeling and Designing Controller of Factors Affecting Profit in Strategic Financial Planning Using Adaptive Neural Control Method
zahra sadeghi mohammad areza motadel abbas toloieMaking more profit in organizations requires them to have accurate tools to strengthen the business for proper financial planning. So far, no comprehensive model for the financial planning of organizations with minimum deviation from predefined goals has been presented. MoreMaking more profit in organizations requires them to have accurate tools to strengthen the business for proper financial planning. So far, no comprehensive model for the financial planning of organizations with minimum deviation from predefined goals has been presented. The purpose of this study is to provide such a model to achieve more accurate management decisions. In this research, an intelligent adaptive control model has been developed using the Elman neural network adaptation algorithm in the system identification process. This model has been developed using data extracted from the official website of the Iranian Stock Exchange and Securities Organization. Some of the most important indicators of profit (as a measure of the company's financial performance) and also optimal profit amount as input and the allowable intervals of its changes to achieve the desired profit as output of the model have been determined and examined. Manuscript profile -
Open Access Article
214 - Comparing the Performance of Linear and Non-Linear Models to Explain Almost Ideal Demand System
Mohammad Rezaei pour Mehdi Zolfaghari mojtaba yousefi dindarloo Abolfazl NajarzadehIn most of empirical studies based on almost ideal demand system (Aids), the elasticity of the price and income estimated by these equations resulted to some sensitive policy making recommendations in microeconomics and macroeconomics. It is in such a case that there is MoreIn most of empirical studies based on almost ideal demand system (Aids), the elasticity of the price and income estimated by these equations resulted to some sensitive policy making recommendations in microeconomics and macroeconomics. It is in such a case that there is some doubt about reliability of linear estimation of such models. In this study, the performance of linear and non-linear almost ideal demand system is under the investigation. For this purpose, seemingly unrelated regression (SURE) method will be applied to estimate linear model and multilayered feed forward neural network (MFNN) is used to estimate a non-linear one. The results indicate that multilayered feed forward neural network is associated with less error than the linear model, and consequently, leads to a better estimation of almost ideal demand system. This result creates some hesitate on application of Stone price index for linear zing estimation of almost ideal demand system. Therefore, it is suggested that feed forward neural network will be applied to estimate almost ideal demand systems. Manuscript profile -
Open Access Article
215 - Stock Price Forecasting through Using ANN and ARIMA Techniques: A Case Study of Pars Petroleum Company
Seyed Nezame aldin Makian Fateme sadat MousaviStock exchange market is one of the important ways to investment. In this market, the investors are looking for the best securities to maximize the profit. Therefore, forecasting the stock price of next day has a vital role in purchasing such securities. To do this, app MoreStock exchange market is one of the important ways to investment. In this market, the investors are looking for the best securities to maximize the profit. Therefore, forecasting the stock price of next day has a vital role in purchasing such securities. To do this, application of Neural Networks financial forecasting has become very popular over the last few years. In this paper, for predicting the next day's close stock price of Pars Petroleum Company, Artificial Neural Network (ANN) and Autoregressive Integrated Moving Average (ARIMA) will be developed, used and compared. The data are daily collected and analyzed during 2009-2011. The findings indicate that forecasting the price by Neural Network is superior to ARIMA due to its less error coefficients and high explanatory ability. Manuscript profile -
Open Access Article
216 - تخمین استحکام کششی قطعات در جوشکاری قوسی تنگستن- گاز با سرعتهای پیشروی کنترل شده با استفاده از شبکه های عصبی
غلامرضا مرامی امیر مصطفی پور اصل رامین مشک آبادی -
Open Access Article
217 - تخمین استحکام فشاری ماسه ریختهگری در مقادیر مختلف رطوبت با استفاده از شبکه عصبی مصنوعی
رامین مشک آبادی غلامرضا مرامی کمال جهانی -
Open Access Article
218 - Investing Neural Network Trianing with Metaheuristic Algorithms in order to Prediction of Iran Stock Index
Seyed Ahmad Mirzaei Zakiyeh Nikdel Zahra NikdelPrediction and analysis of stock market movements are an important topic for researchers, traders and have got an important role in today’s economy. Variety in policies, such as government policies and economic policies affect the stock market and cause stock pric MorePrediction and analysis of stock market movements are an important topic for researchers, traders and have got an important role in today’s economy. Variety in policies, such as government policies and economic policies affect the stock market and cause stock price changes. The predicting stock price movement on a daily basis due to the non-linear and chaotic stock price movements is a difficult task. There are several ways for predicting in stock market. Artificial intelligence techniques have been widely used to predict data with nonlinear and chaotic structure. One of these techniques is neural network. If neural network is trained correctly, then it has minimum error in predicting. In this research, we will train the multi layer perceptron neural network with 8 meta heuristics algorithms and we predict Tehran Exchange Dividend Price Index (TEDPIX). The Results show that grey wolf optimization has the minimum error in training of neural network. Manuscript profile -
Open Access Article
219 - Designing a model for predicting the financial bankruptcy of companies listed in the Tehran Stock Exchange using artificial neural networks and comparing it with the logit regression model
farhad sanchooliConsidering the concerns that investors have about the return of principal and capital gains and the consequences and costs that bankruptcy can cause for companies and the country's economy and other individuals and institutions, the design of a reliable model In order MoreConsidering the concerns that investors have about the return of principal and capital gains and the consequences and costs that bankruptcy can cause for companies and the country's economy and other individuals and institutions, the design of a reliable model In order to predict the probability of bankruptcy of companies, it seems necessary to guide decision makers such as investment companies, banks and the government.In this research, the artificial neural network method and the logit regression method were used to predict the bankruptcy of a number of companies admitted to the Tehran Stock Exchange during the years 2015 to 2019, and the results were compared with the logit regression method. The overall prediction accuracy of the artificial neural network method for each of the years t, t-1, t-2 and t-3 is equal to 96.55%, 96.55%, 92.24% and 24/2 respectively. 92% and for the logit regression method for the same years, it is 94%, 94.82%, 90.51% and 87.06% respectively, which showed that the artificial neural network method has a higher accuracy than the logit regression method. is. Therefore, it can be concluded that the artificial neural network method provides a more appropriate tool for predicting the bankruptcy of companies. Manuscript profile -
Open Access Article
220 - An Intelligent Method for Death Prediction Using Patient Age and Bleeding Volume on CT scan
Yosra Azizi Nasrabadi Ali Jamali Nazari Hamid Ghadiri Farshid Babapour MofradThe purpose of this paper's prediction of survival or death within 30 days is based on a cerebral hemorrhage. Timely and correct diagnosis and treatment of cerebral hemorrhage are essential. If the patient's death is predicted during these thirty days, the treating phys MoreThe purpose of this paper's prediction of survival or death within 30 days is based on a cerebral hemorrhage. Timely and correct diagnosis and treatment of cerebral hemorrhage are essential. If the patient's death is predicted during these thirty days, the treating physician should use intensive care and more treatment for the patient. Cerebral hemorrhages require immediate treatment and rapid and accurate diagnosis. In this article, using the volume of cerebral hemorrhage and the patient's age and using the neural network of support vector machine (SVM), it is predicted what percentage of people with cerebral hemorrhage survive and what percentage die. Parameters of cerebral hemorrhage volume and, age of patients, neural network input are considered. The network's output is the survival or death of patients with cerebral hemorrhage over the next thirty days. The data we used included the bleeding volume and age of 66 patients with lobar hemorrhage, 76 patients with deep bleeding, nine patients with Pontine hemorrhage and 11 patients with cerebellar hemorrhage. All bleeding models are considered as input to the support vector machine neural network. The overall accuracy of the designed support vector machine neural network is 93%. Regardless of the type of cerebral hemorrhage, the survival or death of people with cerebral hemorrhage within 30 days is predicted. Manuscript profile -
Open Access Article
221 - Insulin drug regulation by general type 2 fuzzy controller with alpha plane
Shima Nasr Hamid MahmoodianInsulin therapy with an insulin pump for diabetic patients has different challenges in the real world. Physiological uncertainties in human bodies, different types of daily activities are the most important challenges in this field. Besides, delay in CHO effects in bloo MoreInsulin therapy with an insulin pump for diabetic patients has different challenges in the real world. Physiological uncertainties in human bodies, different types of daily activities are the most important challenges in this field. Besides, delay in CHO effects in blood glucose may increase the risk of hypoglycemic and hyperglycemic. In this paper, general type 2 fuzzy controller with alpha-plane has been used to handle the uncertainties and a neural network predictor to estimate the blood glucose in next hour as well. Genetic algorithm is also used to tune some free parameters in the controller. in addition, Fuzzy rules have been weighted by predefined values based on the prediction of the amount of glucose in one hour late. in such case, rule weighting has been adjusted according to the glucose of the body which in turn two high risk situations of diabetic patients (hyperglycemia and hypoglycemia) have been considered in fuzzy inference. the Simulation results on Hovorka model shows that the controller can regulate the blood glucose in the existence of uncertainty in model and CHO regimen without the risk of hypoglycemic and hyperglycemic situations. Manuscript profile -
Open Access Article
222 - Adaptive Neural Network Dynamic Surface Control for Nonlinear Stochastic Systems in The Strict-Feedback Form with Prandtl-Ishlinskii Hysteresis in The Actuator
Mohammad Mahdi Aghajary Mahnaz HashemiUsing the adaptive radial basis function (RBF) neural network dynamic surface control design method, a controller design approach is presented in order to the stabilization of strict-feedback nonlinear stochastic systems subjected to Prandtl-Ishlinskii nonlinearity in t MoreUsing the adaptive radial basis function (RBF) neural network dynamic surface control design method, a controller design approach is presented in order to the stabilization of strict-feedback nonlinear stochastic systems subjected to Prandtl-Ishlinskii nonlinearity in the actuator. This method is capable to be applied to nonlinear stochastic systems with any unknown dynamics. According to the universal approximation capability the RBF neural networks make it possible to approximate the unknown dynamics of the nonlinear stochastic systems. Using the minimal-learning-parameters algorithm the approximation procedure is done with a minimum complexity and required calculations. The stability of the proposed control system is proven analytically and its results are demonstrated using a simulation example. It is shown that the proposed design approach guarantees the boundedness in probability for adaptive control system, and in turn the uniformly ultimately boundedness of all closed-loop signals. It is also shown, that using this method the tracking error can be made arbitrarily small. Manuscript profile -
Open Access Article
223 - Neural Adaptive Control of an Artificial Pancreas for People with Type 1 Diabetes Under Saturated Insulin Injection Rate
Sadegh Rezaei Mohsen ParsaIt is essential to control vital variables in patients whose natural control system has been compromised for some reason. One of these vital variables is blood glucose levels. Unfortunately, in people with diabetes (blood sugar), blood glucose levels are not regulated p MoreIt is essential to control vital variables in patients whose natural control system has been compromised for some reason. One of these vital variables is blood glucose levels. Unfortunately, in people with diabetes (blood sugar), blood glucose levels are not regulated properly. To compensate for this lack, in recent years, several studies and efforts have been made to build and improve the function of the artificial pancreas to control blood sugar. The presence of factors such as multiple uncertainties due to physiological differences in individuals, various activities during the day, delayed effects of carbohydrates on blood sugar levels, stress and exercise make controlling the artificial pancreas a challenging system. But one of the most important challenges in this area, which has not been less addressed in the literature is the limitation on the allowable dose of insulin injected into the artificial pancreas for patients with type 1 diabetes. On the one hand, injecting a high dose of insulin can cause problems such as hyperglycemia issues and on the other hand, injecting a negative dose of insulin is meaningless. In this paper, after selecting the Bergman model and considering the existence of asymmetric saturation in the actuator, the back-stepping control method is used and it is combined with an adaptive technique to improve the controller performance. Finally, simulation results depict that in the presence of large step disturbance, the insulation rate remains in the allowed band of zero to 20 mU/min, and the blood glucose level does not exceed the appropriate level 130mg/dl. Manuscript profile -
Open Access Article
224 - Evaluation of Deep Neural Networks in Emotion Recognition Using Electroencephalography Signal Patterns
Azin Kermanshahian Mahdi KhezriIn this study, the design of a reliable detection system that is able to identify different emotions with the desired accuracy has been considered. To reach this goal, two different structures for the emotion recognition system include 1) using linear and non-linear fea MoreIn this study, the design of a reliable detection system that is able to identify different emotions with the desired accuracy has been considered. To reach this goal, two different structures for the emotion recognition system include 1) using linear and non-linear features of the electroencephalography (EEG) signal along with common classifiers and 2) using EEG signal in a deep learning structure is considered to identify emotional states. To design the system, the EEG signals of the DEAP database which were recorded by displaying emotional videos from 32 subjects were used. After the preparation and noise removal, linear and non-linear features such as: Skewness, Kurtosis, Hjorth parameters, Lyapunov exponent, Shannon entropy, correlation and fractal dimension and time reversibility were extracted from the alpha, beta and gamma subbands of the EEG signals. Then according to structure 1, the features were applied as input to common classifiers such as decision tree (DT), k nearest neighbor (kNN) and support vector machine (SVM). Also in structure 2, the EEG signal was considered as the input of the convoloutional neural network (CNN). The goal is to evaluate the results of deep learning networks and other methods for emotion recognition. According to the obtained results, the SVM achieved the best performance for identifying four emotional states with 94.1 % accuracy. Also, the proposed CNN identified the desired emotional states with the accuracy of 86%. Deep learning methods are superior to simple classifiers because they do not require the features of the signals and are resistant to different noises. Using a short period of time for the signals and performing near optimal preprocessing and conditioning, can further improve the results of deep neural networks. Manuscript profile -
Open Access Article
225 - Double JPEG Compression Detection Using Spatial-Domain Deep Neural Networks
Mohammad Rahmati Farbod Razzazi Alireza BehradWith the increasing interest in Joint Photographic Experts Group (JPEG) image compression, one of the most important issues in digital image manipulation is finding a proper method to detect double JPEG compression. This paper introduces a trained adaptive filter based MoreWith the increasing interest in Joint Photographic Experts Group (JPEG) image compression, one of the most important issues in digital image manipulation is finding a proper method to detect double JPEG compression. This paper introduces a trained adaptive filter based on spatial-domain convolutional autoencoder (CAE). This filter can remove interference information caused by image content to have a more accurate detection. The convolutional neural network (CNN) has been widely employed for accurate image classification; therefore, a CNN is used in the classification part of the proposed algorithm. The proposed model is based on consecutive CAE with CNN, which is able to provide acceptable detection accuracy and sensitivity to quality factors (QFs) in two scenarios, i.e. aligned and non-aligned forgeries. This model improves the sensitivity to quality factors by up to 86% in the relative error reduction (RER) rate in some cases. Other experiments such as manipulation localization on the RAISE dataset have been performed to evaluate the proposed method. These results show the superior performance of this method compared to similar algorithms in the situations that the quality factor of the second compression is greater the quality factor of the first compression. Manuscript profile -
Open Access Article
226 - Brain Stroke Classification Based on Deep Learning Approach in Microwave Brain Imaging System
Majid Roohi Jalil Mazloum Mohammad Ali Pourmina Behbod GhalamkariOne of the main reasons of death in the world, mostly affecting seniors, is brain stroke. Almost 85% of all brain strokes are ischemic due to internal bleeding in a part of the brain. Due the high mortality rate, quick diagnosic and treatment of ischemic and hemorrhagic MoreOne of the main reasons of death in the world, mostly affecting seniors, is brain stroke. Almost 85% of all brain strokes are ischemic due to internal bleeding in a part of the brain. Due the high mortality rate, quick diagnosic and treatment of ischemic and hemorrhagic strokes are of utmost importance. In this paper, to realize microwave brain imaging system, a circular array-based of modified bowtie antennas located around the multilayer head phantom with a spherical target with radius of 1 cm as intracranial hemorrhage target aresimulated in CST simulator. To obtain satisfied radiation characteristics in the desired band (from 0.5-5 GHz) an appropriate matching medium is designed. First, in the processing section, a confocal image-reconstructing method based using delay and sum (DAS) and delay, multiply and sum (DMAS) beam-forming algorithms is used. The reconstructed images generated shows the usefulness of the proposed confocal method in detecting the spherical target in the range of 1 cm. The main purpose of this paper is stroke classification using deep learning approaches. For this, an image classification algorithm is developed to estimate the stroke type from reconstructed images. By using the proposed deep learning method, the reconstructed images are classified into different categories of cerebrovascular diseases using a multiclass linear support vector machine (SVM) trained with convol­uti­onal neural networks (CNN) features extracted from the images. The simulated results show the suitability of the proposed image reconstruction method for precisely localizing bleeding targets, with 89% accuracy in 9 seconds. In addition, the proposed deep-learning approach shows good performance in terms of classification, since the system does not confuse between different classes. Manuscript profile -
Open Access Article
227 - Fire Detection Based on Extraction of Spatio-Temporal Features by Convolutional Neural Networks and Fractal Analysis
Monir Torabian Hossein Pourghassem Homayoun Mahdavi-Nasab Payam SanaeeFire is one of the dangers that can endanger human health in a short time and if it is not controlled in time, it will cause a lot of damage. Therefore, timely and accurate identification of the location of the fire can prevent the consequences of its expansion. In this MoreFire is one of the dangers that can endanger human health in a short time and if it is not controlled in time, it will cause a lot of damage. Therefore, timely and accurate identification of the location of the fire can prevent the consequences of its expansion. In this research, a new method for fire detection is proposed based on the extraction of its temporal-spatial features in video frames. In the proposed algorithm, a multiscale convolutional neural network along with a YOLO (you only look once) network is used to extract spatial features and identify fire candidate regions. Then, fractal analysis based on the temporal blanket method is then used to remove non-moving textures similar to fire and to examine the temporal features of the candidate region. Finally, the fire region is separated from the other parts of the image by fusion the results of the two steps. The evaluation results of the proposed method on three data sets show that the accuracy of fire detection is about 96.1%, while the precision and recall values are 92% and 96.9%, respectively. Experimental results show that the proposed method performs better than existing algorithms and thus confirms the ability of this method for efficient use in the real world. Manuscript profile -
Open Access Article
228 - Load Frequency Control in a Hybrid Power System Considering Renewable Energy Sources and Electric Vehicles Using Fractional Order PID Controller Based on Wavelet Neural Network
Abbas-Ali Zamani Seyed Mohammad Kargar Dehnavi Alireza ReisiRestructuring of power systems and integration of different renewable energy sources with complex dynamic behaviors and high structural uncertainties has made the issue of load frequency control more important. For a hybrid power system that includes a thermal power pla MoreRestructuring of power systems and integration of different renewable energy sources with complex dynamic behaviors and high structural uncertainties has made the issue of load frequency control more important. For a hybrid power system that includes a thermal power plant taking into account nonlinear limitations such as the governor dead band and generator rate constraints and renewable energy sources including a wind turbine, solar-thermal power plant, electrolyzer, fuel cell, and plug-in electric vehicle, this paper proposes an adaptive wavelet neural network fractional order PID controller (AWNNFOPID) based on self-recursive wavelet neural networks and fractional order PID controller. To compare the performance of the proposed AWNNFOPID controller, four different scenarios are considered and the simulation results are compared with traditional I, PI, and PID controllers as well as with the optimized FOPID controller. The simulation results show that the proposed AWNNFOPID controller has better performances than the other control strategies used for the studied hybrid power system based on performance indicators such as settling time, rise time, maximum overshoot, maximum undershoot, integral time absolute error (ITAE), and integral absolute error (IAE). Manuscript profile -
Open Access Article
229 - Improvement and Optimization of Homogeneous Composite in Array Antennas using Convolution Neural Network
Gohar Varamini Behnam Dorostkar YaghoutiAntenna structure and performance, bandwidth, gain and guidance are the most important performance indicators. For this purpose, RL homogeneous transmission line is very important due to low loss, phase changes, frequency bandwidth, zero and negative order resonance, mi MoreAntenna structure and performance, bandwidth, gain and guidance are the most important performance indicators. For this purpose, RL homogeneous transmission line is very important due to low loss, phase changes, frequency bandwidth, zero and negative order resonance, miniaturization and easy construction, and is very suitable in the design of broadband and array antennas. The right-left hand structure in the antennas due to the difference in the phase of the right part in the repetition of arrays and the thickness of the layer has phase delay and finally deviation of the radiation pattern. On the other hand, the blockage of the transmission line on the left causes bandwidth restriction and increasing the number of casualties in the system. In this paper, with the help of deep learning (DL), composite defects are solved and optimized arrayed antenna. The proposed antenna transmission line design in the range of 2-7 GHz, optimum resonance frequency of 4.5 GHz and convolution, dual resonance and spiral inductor neural algorithm are loaded onto the patch in four arrays. The use of convolutional neural network (CNN) in the left transmission line compensates for the right phase delay and finally enables optimal phase changes, correction of radiation pattern and continuous scanning of phase arrays. Also, by creating gaps in the microstrip patch, bandwidth limit is removed and the system losses are reduced. Secondary dimensions compared to the primary dimension are reduced to about 60% in size and miniature according to the smart modified model. The results of this improved composite showed an increase in bandwidth of 20.3 and the efficiency of the radiation pattern by more than 96%. On the other hand, small dimensions, appropriate frequency bandwidth and simple network design have been provided. Manuscript profile -
Open Access Article
230 - Reduction of Sub-synchronous Resonances with D-FACTS Devices using intelligent Control ,
Zahra Amini Abbas KargarWhen a turbine–generator set connect to a long transmission line, may results side effects such as Sub-Synchronous Resonances (SSR). The capabilities of the Distributed Static Series Compensator (DSSC) as a member of the family of D-FACTS can be used to reduce the MoreWhen a turbine–generator set connect to a long transmission line, may results side effects such as Sub-Synchronous Resonances (SSR). The capabilities of the Distributed Static Series Compensator (DSSC) as a member of the family of D-FACTS can be used to reduce these SSR. To achieve this desired goal, the fuzzy controller, Particle Swarm Optimization (PSO) and artificial neural network is used to control of the DSSC. Particle swarm optimization is designed Based on the Conventional Damping Controller (CDC) and fuzzy logic is designed based on damping controller (FLBDC) and damping control based on artificial neural network trained using the fast pace of changes has been designed. Stability of the system is analysed by simulations in the time domain with performance index (PI). All simulations are done using Matlab / Simulink software. Case studies show that proposed algorithms can reduce SSR in the system.All simulations are done using Matlab / Simulink software. Case studies show that proposed algorithms can reduce SSR in the system. Manuscript profile -
Open Access Article
231 - Decentralized Adaptive Control of Large-Scale Non-Affine Nonlinear Time-Delay Systems Using Wavelet Neural Networks
Elaheh Saeedi Bahram Karimi Mostafa \خخقذثاهIn this paper, a decentralized adaptive controller with using wavelet neural network is used for a class of large-scale nonlinear systems with time- delay unknown nonlinear non- affine subsystems. The entered interruptions in subsystems are considered nonlinear with tim MoreIn this paper, a decentralized adaptive controller with using wavelet neural network is used for a class of large-scale nonlinear systems with time- delay unknown nonlinear non- affine subsystems. The entered interruptions in subsystems are considered nonlinear with time delay, this is closer the reality, compared with the case in which the delay is not considered for interruptions. In this paper, the output weights of wavelet neural network and the other parameters of wavelet are adjusted online. The stability of close loop system is guaranteed with using the Lyapanov- Krasovskii method. Moreover the stability of close loop systems, guaranteed tracking error is converging to neighborhood zero and also all of the signals in the close loop system are bounded. Finally, the proposed method, simulated and applied for the control of two inverted pendulums that connected by a spring and the computer results, show that the efficiency of suggested method in this paper. Manuscript profile -
Open Access Article
232 - Evaluation of the Performance of Feedforward and Recurrent Neural Networks in Active Cancellation of Sound Noise
Mehrshad Salmasi Homayoun Mahdavi-NasabActive noise control is based on the destructive interference between the primary noise and generated noise from the secondary source. An antinoise of equal amplitude and opposite phase is generated and combined with the primary noise. In this paper, performance of the MoreActive noise control is based on the destructive interference between the primary noise and generated noise from the secondary source. An antinoise of equal amplitude and opposite phase is generated and combined with the primary noise. In this paper, performance of the neural networks is evaluated in active cancellation of sound noise. For this reason, feedforward and recurrent neural networks are designed and trained. After training, performance of the feedforwrad and recurrent networks in noise attenuation are compared. We use Elman network as a recurrent neural network. For simulations, noise signals from a SPIB database are used. In order to compare the networks appropriately, equal number of layers and neurons are considered for the networks. Moreover, training and test samples are similar. Simulation results show that feedforward and recurrent neural networks present good performance in noise cancellation. As it is seen, the ability of recurrent neural network in noise attenuation is better than feedforward network. Manuscript profile -
Open Access Article
233 - Partial Discharge Analysis in Generator Stator Winding Using Artificial Neural Network
Seyed Meisam Ezzati Faramarz Faghihi Ali Moarefian poorThis paper analyses the Partial Discharge (PD) on the stator terminals of synchronous generator. This is necessary to collect experimental data for the analysis. First, exploiting the measurement devices, special signals that describe the partial discharges are repeated MoreThis paper analyses the Partial Discharge (PD) on the stator terminals of synchronous generator. This is necessary to collect experimental data for the analysis. First, exploiting the measurement devices, special signals that describe the partial discharges are repeatedly collected. Then, based on the current standards, the collected empirical data are subjected to interpretation. To ease the interpretation process, an Artificial Neural Network is trained and validated. We have used a double layers forward perceptron neural network which is trained by Levenberg–Marquardt algorithm that utilizes least square method as the performance index. As the case study, three gas turbine-generators located in Shahre-Rey power plant (Rey Power Generation Management Company) have been subjected to repeatedly data collection. The mentioned generators are manufactured by Mitsubishi with 85 MW of nominal power. Generally, partial discharge analysis has the following practical implication about the probable defects: lamination of the internal terminal, mobility within the main insulation and discharge into the groove in stator of synchronous generator. Manuscript profile -
Open Access Article
234 - Coordinated Control of FACTS Devices by Using ADALINE Neural Network to Enhance the Transient Stability of Power System
Mehdi Ghasemi Ali Reza Roosta Bahador FaniIn order to improve transient stability and increase the system damping, this paper introduces a specific way of coordination between FACTS devices (TCSC and SVC). In order to improve the performance and use all the features of TCSC and SVC (presented in this article), MoreIn order to improve transient stability and increase the system damping, this paper introduces a specific way of coordination between FACTS devices (TCSC and SVC). In order to improve the performance and use all the features of TCSC and SVC (presented in this article), it is necessary a controller be used in which does not have the limitations of other controllers and simply be also able to quickly respond and adapt to the power system model. Hence, these features can be found in the intelligent controllers in which the ADALINE network is one of them. To better understand the performance of the ADALINE network controller; this controller will be compared with a controller which is designed by the optimal control parameters (LQR). The instruments used for FACTS are from the injection type and therefore, it is possible to use a fixed factorization ybus matrix in the calculations. Simulation results using non-linear network show that the ADALINE neural network controller has better performance than the LQR controller and can cause significant improvement on damping and transmission ability in the power system. Manuscript profile -
Open Access Article
235 - Automatic Persian License Plate Recognition by Edge Detection Using Hopfield Neural Network
Homayoun Mahdavi-Nasab Mohammad Sadegh Memarzadeh Peyman MoallemLicense plate is the most suitable information for automobile identification. Auto license plate recognition system is an automatic process which extracts the plate number using photographs of the auto. The method presented in this paper consists of two stages. First, t MoreLicense plate is the most suitable information for automobile identification. Auto license plate recognition system is an automatic process which extracts the plate number using photographs of the auto. The method presented in this paper consists of two stages. First, the plate is located by edge detection and morphological techniques. Second, the characters are identified using Hopfield neural network. The proposed method has been tested on 700 photos with different backgrounds, distances and angles. The correct plate location and identification are evaluated 97.8% and 93% respectively. Manuscript profile -
Open Access Article
236 - New Prognostic Index to Detect the Severity of Asthma Automatically Using Signal Processing Techniques of Capnogram
Mohsen Kazemi Aik Howe TeoIn this paper, a new prognostic index to detect the severity of asthma by processing capnogram signals is presented. Previous studies have shown significant correlation between the capnogram and asthmatic patient. However, most of them used conventional time-domain meth MoreIn this paper, a new prognostic index to detect the severity of asthma by processing capnogram signals is presented. Previous studies have shown significant correlation between the capnogram and asthmatic patient. However, most of them used conventional time-domain methods and based on assumption that the capnogram is a stationary signal. In this study, by using linear predictive coding (LPC) coefficients and autoregressive (AR) modelling (Burg method), the capnogram signals are processed. Then, a number of six features including α1, and α4 from LPC and power spectral density (PSD) parameters through AR modelling are extracted. After that, by means of receiver operating characteristic (ROC) curve, the effectiveness of the extracted features to differentiate between asthmatic and nonasthmatic conditions is justified. Finally, selected features are used in a Gaussian radial basis function (GRBF) network. The output of this network is an integer prognostic index ranging from 1 to 10 (depends on the severity of asthma) with an average good detection rate of 90.15% and an error rate of 9.85%. In the other word, based on the results, sensitivity and specificity of this algorithm are 93.54% and 98.29%, respectively. This developed algorithm is purposed to provide a fast and low-cost diagnostic system to help healthcare professional involved in respiratory care as it would be possible to monitor severity of asthma automatically and instantaneously. Manuscript profile -
Open Access Article
237 - System Identification of a Nonlinear Multivariable Steam Generator Power Plant Using Time Delay and Wavelet Neural Networks
Laila Khalilzadeh Ganjali-khani Farid Sheikholeslam Homayoun Mahdavi-NasabOne of the most effective strategies for steam generator efficiency enhancement is to improve the control system. For such an improvement, it is essential to have an accurate model for the steam generator of power plant. In this paper, an industrial steam generator is c MoreOne of the most effective strategies for steam generator efficiency enhancement is to improve the control system. For such an improvement, it is essential to have an accurate model for the steam generator of power plant. In this paper, an industrial steam generator is considered as a nonlinear multivariable system for identification. An important step in nonlinear system identification is the development of a nonlinear model. In recent years, artificial neural networks have been successfully used for identification of nonlinear systems in many researches. Wavelet neural networks (WNNs) also are used as a powerful tool for nonlinear system identification. In this paper we present a time delay neural network model and a WNN model in order to identify an industrial steam generator. Simulation results show the effectiveness of the proposed models in the system identification and demonstrate that the WNN model is more precise to estimate the plant outputs. Manuscript profile -
Open Access Article
238 - Predicting Hook -Shaped and Concrete Steel Fibers Adhesion Parameters Using Artificial Neural Networks
amir ebrahim akbari baghal Amir ebrahim akbari bagalGiven the importance of using steel fibers in reinforcement of concrete, in the present study using artificial neural networks to predict the behavior of hook -shaped steel fibers from concrete. Due to the constraint of comprehensive laboratory data, data obtained from MoreGiven the importance of using steel fibers in reinforcement of concrete, in the present study using artificial neural networks to predict the behavior of hook -shaped steel fibers from concrete. Due to the constraint of comprehensive laboratory data, data obtained from limited element analysis has been used as neural network input. The fibers are used to simulate the fiber and the Abacus software. In the limited element model, the interactions between fibers and concrete are simulated using the concept of the transitional area of the common surface whose parameters are obtained using the reversed limited element method and the use of the out -of -the -way experimental test results on a fiber sample. After assessment of the numerical model results with the empirical results, the results were extracted for effective parameters of the fibers and based on them using neural networks. Forecasting of the outburst has been carried out by the Multi-Layer Artificial Neural Networks and the Rear Publishing Algorithm, with Marcoradet-Clberg optimization techniques. The results show that the neural network model presented in this study, due to the ability to use more variables in modeling and more accurate results, is an effective way to predict the fiber's extrusion force. Manuscript profile -
Open Access Article
239 - Online adaptive neuro-fuzzy controller design to attenuate the seismic responses in a 20-story benchmark structure
Rasoul Sabetahd Seyed Arash Mousavi Ghasemi Ramin Vafaei Poursorkhabi Ardashir Mohammadzadeh Yousef ZandiIn the present research, design of a strong and online adaptive controller in the active cable control system is discussed to overcome the earthquake vibrations of multi-story buildings. Considering all variables as unknown, this study introduces a new type 2 adaptive n MoreIn the present research, design of a strong and online adaptive controller in the active cable control system is discussed to overcome the earthquake vibrations of multi-story buildings. Considering all variables as unknown, this study introduces a new type 2 adaptive neuro-fuzzy controller. Using the MLP neural network (multi-layer perceptrons), Jacobian and the structural system estimation are extracted. This estimated structural system model is implemented into the online controller system in the next step. Adaptive controllers are tuned using a post-propagation algorithm and Extended Kalman Filter and are thus able to control and tune the controllers and the cable system. In this method, a PID controller is also used, which increases the strength and stability of the adaptive neural-fuzzy controller system two against earthquake vibrations. The superiority of the proposed controller system over an online simple adaptive controller is also demonstrated. This controller is utilized as an implicit reference model. In this proposed method, Extended Kalman Filter is innovatively used to tune online controllers. In this research, the performance of both controllers is investigated under the far and near fault field pressures. Based on the numerical results, the adaptive neural-fuzzy controller performs about 21% better than the online simple adaptive controller in minimizing the seismic responses of the structure during an earthquake and reaching the control criteria when the parametric characteristics of the structure change. Manuscript profile -
Open Access Article
240 - Adaptive Control of the 3-Story Benchmark Building Equipped with MR Damper using Fractional Order Robust Controller
Ommegolsoum Jafarzadeh Seyed Arash Mousavi Ghasemi seyyed Mehdi Zahraei Ardashir Mohammadzadeh Ramin Vafaei PoursorkhabiThe goal of the present research is to propose a novel adaptive fractional order PID (AFOPID) controller whose parameters are tuned online by five exclusive multilayer perceptron (MLP) neural networks using the extended Kalman filter (EKF). An MLP neural network that is MoreThe goal of the present research is to propose a novel adaptive fractional order PID (AFOPID) controller whose parameters are tuned online by five exclusive multilayer perceptron (MLP) neural networks using the extended Kalman filter (EKF). An MLP neural network that is trained using the Back Propagation (BP) error algorithm is considered to identify the structural system and estimate the plant. The Jacobian of the model estimated online is utilized to apply to the controller. Considering the adaptive interval type-2 fuzzy neural networks (IT2FNN) and this issue that the compensator is tunned by EKF and feedback error learning strategy (FEL), the stability and robustness of this controller are increased against the estimation error, seismic disturbances, and some unknown nonlinear functions. In order to validate, the performance of the proposed controller is investigated on a 3-story nonlinear benchmark building equipped with semi-active dampers under far and near field earthquakes. In order to evaluate the effectiveness of the proposed controller equipped with a compensator in reducing seismic responses, the evaluation indices were discussed and compared with previous studies. The numerical results represent the substantial efficiency of the proposed adaptive controller (AFOPID) over the previous controllers such that J2 in the Hachinohe and Northridge earthquakes enhanced by up to 35% and more than 40%, respectively. In general, all indices ( J3 to J6 ) have experienced a considerable enhancement using the proposed method. Manuscript profile -
Open Access Article
241 - Production of Synthetic Seismic Records Using Fuzzy Neural Network
Peyman Shadman Mehdi Amri Mohammad KhorasaniThere is a growing need for dynamic time history analysis and the absence of proper records in different areas has necessitatedthe production of artificiallaccelerograms compatible with the whole plan. This study presents a new approachbased on wavelet packet transform MoreThere is a growing need for dynamic time history analysis and the absence of proper records in different areas has necessitatedthe production of artificiallaccelerograms compatible with the whole plan. This study presents a new approachbased on wavelet packet transform and artificial intelligence techniques to produce artificial earthquake accelerograms compatible with the whole plan. This approachtakes into account the magnitude and the distance from the fault. The study of neural networks and fuzzy wavelet packet analysis has been used to achieve the desired goal. To do so, first earthquake accelerograms have been collected according to specific site conditions, earthquake magnitude and distance from origin.Then all records have been gatehered for training with fuzzy neural network. Attenuation spectra have been developed on the basis of information in the area using nonlinear regression. Then using fuzzy neural networks, the relationship between earthquake records and the devloped spectra from each record is calculated. In this satge, using wavelet packet analysis, mapping acceleration are analyzed and converted intoaccelerograms (wavelet coefficients) Manuscript profile -
Open Access Article
242 - Predication of Some Physiochemical Properties of Low Calorie Cake Containing Apple Fiber Using Artificial Neural Networks
Maryam Sabet Ghadam Mohammad Reza Saeedi Asl Akram Sharifi Ahmad Pedram Nia Mohammad Armin -
Open Access Article
243 - بهینهسازی فرایند استخراج روغن از دانههای گلرنگ با پیشتیمار مایکروویو و تخمین پارامترهای فرایند با کمک شبکهی عصبی مصنوعی
زهرا دولت آبادی سیدعلی مرتضوی معصومه مقیمی حمید بخش آبادی سید حسین استیری -
Open Access Article
244 - مدلسازی استخراج روغن از دانه کتان با پیش تیمار میدان الکتریکی پالسی با استفاده از شبکه عصبی مصنوعی
شکوفه غراوی مسعود بذرافشان معصومه مقیمی -
Open Access Article
245 - پیشگویی فعالیت رادیکالگیرندگی، شمارش آغازگرها و خواص حسی ماست پروبیوتیک حاوی عصارههای هیدروالکلی اسپیرولینا پلاتنسیس و گیاه چویل با شبکه عصبی مصنوعی
عبد الرضا آقاجانی سید علی مرتضوی فریده طباطبایی یزدی -
Open Access Article
246 - بهینهسازی فرآیند آبگیری اسمزی شلیل و مدلسازی پارامترهای آبگیری با کمک شبکهی عصبی مصنوعی
حمید بخش آبادی معصومه مقیمی زهرا دولت آبادی سحر اصغری پور -
Open Access Article
247 - مدلسازی پراکنده شدن ذرات فیتواسترول در امولسیون روغن/آب با استفاده از شبکه عصبی و رگرسیون چند متغیره
زهرا ایزدی علی نصیرپور محبوبه استادزاده -
Open Access Article
248 - بهینهسازی فرمولاسیون نان حجیم بدون گلوتن ذرت حاوی کنسانتره پروتئین آب پنیر و آنزیم ترانسگلوتامیناز میکروبی
نساء صفوی مهدی قره خانی -
Open Access Article
249 - مدل ترکیبی شبکهی عصبی و تحلیل پوششی داده ها برای ارزیابی کارایی عملکرد واحدها
صادق حیدری احسان زنبوری حمید پروینکایی و ارزیابی یکی از اصلیترین و مهمترین نیاز های سازمانها، شرکتها و موسسات میباشد و این سازمان ها چون با حجم زیادی از داده سر و کار دارند. تحلیل پوششی دادهها روشی مناسب برای کارایی و ارزیابی عملکرد سازمانها میباشد. این تحقیق برای ارزیابی عملکرد و کارایی واحدهای Moreکایی و ارزیابی یکی از اصلیترین و مهمترین نیاز های سازمانها، شرکتها و موسسات میباشد و این سازمان ها چون با حجم زیادی از داده سر و کار دارند. تحلیل پوششی دادهها روشی مناسب برای کارایی و ارزیابی عملکرد سازمانها میباشد. این تحقیق برای ارزیابی عملکرد و کارایی واحدهای تصمیم گیرنده انجام گرفته است، ابتدا رویکردی با مدل BCC خروجی محور رتبهبندی واحدهای کارا در قالب مدلهای تحلیل پوششی دادهها مورد بررسی قرار گرفت و ضعف مدل، از نظر محاسبه و تفکیک کارایی مشخص گردید سپس برای از بین رفتن این مشکلات از روش ترکیبی تحلیل پوششی دادهها مدل BCC خروجی محور و شبکه عصبی مصنوعی به منظور ارزیابی کارایی این واحدها استفاده گردید تا بتوان این مشکل را بر طرف نمود. در پایان نیز مقایسهای بین نتایج حاصل از دو مدل انجام گرفته است. با توجه به مقدار کارایی بدست آمده با روش bcc خروجی محور، مشاهده می گردد تعدادی از واحدها مقدار کارایی آنها برابر با یک است که این باعث میگردد نتوانیم این واحدها رتبه بندی نماییم. اما با استفاده از روش پیشنهادی Neuro-DEA هیچ دو شعبه ای دارای مقدار کارایی برابر نبوده و با توجه به کارایی بدست آمده به راحتی می توان این واحد ها را ارزیابی و رتبه بندی نمود. Manuscript profile -
Open Access Article
250 - یک مدل شبکه عصبی بازگشتی برای حل مدل CCR در تحلیل پوششی داده ها
معصومه عباسی عباس قماشیدر این مقاله ما یک شبکه عصبی برای حل مدل CCR در تحلیل پوششی داده ها (DEA) معرفی می کنیم. مدل شبکه عصبی پیشنهادی از یک مسئله مینیمم سازی نامقید حاصل می شود. از دیدگاه تئوری، نشان داده می شود که شبکه عصبی پیشنهادی به مفهوم لیاپانف پایدار و همگرای عمومی به جواب بهینه مدل C Moreدر این مقاله ما یک شبکه عصبی برای حل مدل CCR در تحلیل پوششی داده ها (DEA) معرفی می کنیم. مدل شبکه عصبی پیشنهادی از یک مسئله مینیمم سازی نامقید حاصل می شود. از دیدگاه تئوری، نشان داده می شود که شبکه عصبی پیشنهادی به مفهوم لیاپانف پایدار و همگرای عمومی به جواب بهینه مدل CCR می باشد. مدل پیشنهادی ساختار تک لایه دارد. با یک مثال عددی موثر بودن مدل پیشنهادی برای حل مدل CCR در DEA نشان داده می شود. Manuscript profile -
Open Access Article
251 - تعیین اندازه گل و رنگ پوست بره های زندی با استفاده از پردازش تصویر و شبکه عصبی مصنوعی
م. خجسته کی ع.ا. اسلمی نژاد ع.ر. جعفری اروریدر این مطالعه، روشی بر مبنای استفاده از پردازش تصویر و شبکه عصبی مصنوعی برای تعیین رنگ و نوع گل پوست در بره ­های نوزاد گوسفند زندی معرفی شده است. داده­ ها از 300 بره­ نوزاد در مرکز پرورش گوسفند زندی خجیر تهران جمع ­آوری شد. در ابتدا، اندازه و شکل گل پوست Moreدر این مطالعه، روشی بر مبنای استفاده از پردازش تصویر و شبکه عصبی مصنوعی برای تعیین رنگ و نوع گل پوست در بره ­های نوزاد گوسفند زندی معرفی شده است. داده­ ها از 300 بره­ نوزاد در مرکز پرورش گوسفند زندی خجیر تهران جمع ­آوری شد. در ابتدا، اندازه و شکل گل پوست بره ­های تازه متولد شده توسط ارزیاب ­های با تجربه ثبت شد و به طور هم­زمان، چندین عکس دیجیتال از نمای جانبی هر بره گرفته شد. ویژگی­ های مربوط به اندازه گل و رنگ پوست بره­ ها از تصاویر دیجیتال با استفاده از ابزار پردازش تصویر (IPT) نرم­ افزار MATLAB استخراج شد. برای تعیین رنگ پوست، طبقه ­بندی پوست براساس اندازه گل و نیز برای برآورد اندازه گل پوست بره ­ها سه شبکه عصبی مصنوعی مجزا طراحی شد. رنگ پوست بره ­ها با استفاده از شبکه عصبی مصنوعی با دقت 100 درصد تعیین شد. دقت شبکه عصبی آموزش ­دیده برای طبقه­ بندی پوست بره ­ها بر اساس اندازه گل آنها 87/94 درصد بود. همچنین دقت شبکه عصبی سوم برای برآورد اندازه گل­ های پوست 44/98 درصد بود. همبستگی بین اندازه گل برآورد شده با استفاده از شبکه عصبی مصنوعی و اندازه گل تعیین شده توسط ارزیاب 4/96 درصد (0.01>P) بود. نتایج این مطالعه نشان داد که امکان استفاده از هوش مصنوعی به عنوان جایگزین ارزیابی انسانی در ثبت صفات پوست وجود دارد. Manuscript profile -
Open Access Article
252 - کاربرد مدل خطی و شبکه عصبی مصنوعی برای پیشبینی عملکرد رشد در جوجههای گوشتی
ش. غضنفریاین مطالعه به منظور پیشبینی عملکرد رشد با استفاده از مدل خطی و شبکه عصبی مصنوعی در جوجههای گوشتی انجام شد. شبکه عصبی مصنوعی ابزار قدرتمندی برای سیستم مدلینگ در دامنه وسیعی از کاربردها است. مدل شبکه عصبی مصنوعی با الگوریتم پس انتشار به طور موفقیت آمیزی ارتباط بین ورودی Moreاین مطالعه به منظور پیشبینی عملکرد رشد با استفاده از مدل خطی و شبکه عصبی مصنوعی در جوجههای گوشتی انجام شد. شبکه عصبی مصنوعی ابزار قدرتمندی برای سیستم مدلینگ در دامنه وسیعی از کاربردها است. مدل شبکه عصبی مصنوعی با الگوریتم پس انتشار به طور موفقیت آمیزی ارتباط بین ورودیها (انرژی قابل سوخت و ساز (کیلوکالری/کیلوگرم) و پروتئین خام (گرم/کیلوگرم) و خروجیها (مصرف خوراک، افزایش وزن و ضریب تبدیل خوراک) را آموزش داد. ارزش R2و T بالا برای مدل شبکه عصبی مصنوعی در مقایسه با مدل خطی نشان داد که شبکه عصبی مصنوعی یک روش مؤثر برای پیشبینی عملکرد رشد در دوره آغازین برای جوجههای گوشتی است. همچنین، گسترش آزمایش با سطوح بیشتری از ورودیها برای پیشبینی عملکرد با استفاده از بهترین مدل شبکه عصبی مصنوعی انجام شد. Manuscript profile -
Open Access Article
253 - مقایسه شبکه عصبی مصنوعی و مدلهای رگرسیونی برای پیشبینی وزن بدن در بز کرکی راینی
م. خورشیدی-جلالی م.ر. محمدآبادی ع. اسمعیلیزاده ا. برازنده ُ.ا. بابنکوشبکه­های عصبی مصنوعی الگوریتم­های آموزشی و مدل­های ریاضی هستند که توانایی تقلید از مغز انسان در پردازش اطلاعات را دارند و می­توانند داده­های پیچیده و غیر خطی را مورد استفاده قرار دهند. هدف این پژوهش مقایسه شبکه عصبی مصنوعی و مدل­های رگرسیونی برای Moreشبکه­های عصبی مصنوعی الگوریتم­های آموزشی و مدل­های ریاضی هستند که توانایی تقلید از مغز انسان در پردازش اطلاعات را دارند و می­توانند داده­های پیچیده و غیر خطی را مورد استفاده قرار دهند. هدف این پژوهش مقایسه شبکه عصبی مصنوعی و مدل­های رگرسیونی برای پیشبینی وزن بدن در بز کرکی راینی بود. داده­های 1389 بز برای وزن بدن، ارتفاع جدوگاه، طول بدن و قفسه سینه مورد استفاده قرار گرفت. مدل­های رگرسیونی مختلف با تمام فاکتورهای ثابت برای بیشتر حالت­های ممکن و با درجه­های مختلف محاسبه شدند و دو شبکه عصبی مصنوعی با لایه­های مخفی متفاوت، توابع آموزش و توابع انتقال گوناگون استفاده شدند. در نهایت، مدل پرسپترون چند لایه با یک لایه مخفی به همراه نرون­ها انتخاب و استفاده شد. همبستگی بین وزن بدن و اندازهگیری­هایش نشان داد که می­توان از اندازه­های بدن برای پیشبینی وزن بدن استفاده کرد و هرچه اندازه­های بیشتری استفاده شوند پیشبینی دقیق­تری انجام خواهد شد. براساس پارامترهای R2و MSE، بهترین معادله رگرسیون فیت شده برای پیشبینی وزن بدن با استفاده از اندازهگیری­های ابعاد بدن انتخاب شد. در حالیکه هر سه اندازه در مدل اثر معنیداری داشتند (0001/0P<)، ارتفاع جدوگاه بالاترین ضریب را داشت (65/0)، بنابراین می­تواند بیشترین اثر را در پیشبینی داشته باشد. مقایسه دو مدل نشان داد که هر دو مدل می­توانند به خوبی وزن بدن را، نزدیک به وزن واقعی آن پیشبینی کنند، اما توانایی شبکه عصبی مصنوعی بالاتر است (R2 برای شبکه عصبی مصنوعی 86/0 و برای مدل­های رگرسیونی 76/0) و به ورن واقعی بدن نزدیک­تر می­باشد. با این وجود، اگر اندازه­های مرتبط بیشتری رکوردبرداری شوند می­توان نتایج مطلوب­تری را با شبکه عصبی مصنوعی به دست آورد. بنابراین، از شبکه عصبی مصنوعی می­توان به جای روش­های سنتی مرسوم برای پیشبینی وزن واقعی بدن با استفاده از اندازه­های بدن استفاده کرد. Manuscript profile -
Open Access Article
254 - کاربرد مدلهای ریاضی برای تخمین میزان انرژی قابل متابولیسم اقلام خوراکی انرژیزا در طیور
M. Sedghi K. Tayebipoor B. Poursina M. Eman Toosi P. Soleimani Roudiاین مطالعه با هدف بررسی امکان پیشبینی انرژی قابل متابولیسم ظاهری (AME) در طیور با استفاده از آنالیز تقریبی در نمونههای گندم (51 نمونه)، جو (56 نمونه) و یولاف (34 نمونه) انجام شد. از آنالیز رگرسیون گام به گام برای ارزیابی رابطه بین AME با نشاسته، عصاره اتری (EE)، فیبر Moreاین مطالعه با هدف بررسی امکان پیشبینی انرژی قابل متابولیسم ظاهری (AME) در طیور با استفاده از آنالیز تقریبی در نمونههای گندم (51 نمونه)، جو (56 نمونه) و یولاف (34 نمونه) انجام شد. از آنالیز رگرسیون گام به گام برای ارزیابی رابطه بین AME با نشاسته، عصاره اتری (EE)، فیبر خام (CF)، قندهای محلول (SS)، خاکستر (ash) و پروتئین خام (CP) در نمونههای گندم و جو، و ماده خشک (DM) ، CF، Ash و CP در نمونههای یولاف استفاده شد. براساس نتایج رگرسیون گام به گام، SS، CF و Ash، بهترین متغیرهای ورودی برای پیشبینی AME در نمونههای گندم بودند. بعلاوه، نشاسته،CF و EE در نمونه های جو و CF و CP در نمونههای یولاف بهترین پارامترهای ورودی جهت تخمین AME شناسایی شدند. همچنین برای بدست آوردن بهترین مدل تخمین زننده AME در این غلات با استفاده از ورودیهای انتخاب شده از رگرسیون خطی چندگانه (MLR) و شبکه عصبی مصنوعی (ANN) استفاده شد. نتایج MLR و ANN نیز نشان داد که SS، CF و ash فاکتورهای مؤثری جهت تخمین AME درگندم میباشند. برای پیشبینی AME نمونههای جو CF، EE و نشاسته متغییرهای مستقل خوبی به شمار میآیند. همچنین CF و CP پارامترهای خوبی برای پیشبینی AME در نمونههای یولاف میباشند. در ارتباط با کارایی مدلها، دقت مدل ANN بالاتر از مدل MLR بود. بر اساس این نتایج، میتوان چنین نتیجهگیری کرد که استفاده از ترکیبات شیمیایی موجود در نمونههای گندم، جو و یولاف، همراه با مدل ANN، روش کاربردی جهت پیشبینی AME این غلات در تغذیه طیور میباشد. Manuscript profile -
Open Access Article
255 - مقایسه کارآیی شبکه عصبی مصنوعی و رگرسیون چندگانه در پیشبینی وزن دنبه گوسفند
م.ع. نوروزیان م. وکیلی علویجهدر این مطالعه ارتباط بین وزن­های تولد، از شیرگیری و پایان پروار با وزن دنبه 69 رأس گوسفند بلوچی توسط روش­های شبکه عصبی مصنوعی و رگرسیون چندگانه بررسی شد. هر دو روش با دقت بالایی وزن دنبه را پیش­بینی کردند. هر چند که میانگین خطا به صورت معنی­داری در روش ش Moreدر این مطالعه ارتباط بین وزن­های تولد، از شیرگیری و پایان پروار با وزن دنبه 69 رأس گوسفند بلوچی توسط روش­های شبکه عصبی مصنوعی و رگرسیون چندگانه بررسی شد. هر دو روش با دقت بالایی وزن دنبه را پیش­بینی کردند. هر چند که میانگین خطا به صورت معنی­داری در روش شبکه عصبی مصنوعی کمتر از رگرسیون چندگانه بود. ضریب تعیین برآورد شده در روش شبکه عصبی مصنوعی (93/0) بالاتر از رگرسیون چندگانه (81/0) به دست آمد. استفاده از شبکه عصبی مصنوعی میانگین خطای استاندارد را 59 و ضریب تعیین را 15 درصد بهبود داد. به نظر می­رسد که بتوان با استفاده از شبکه عصبی مصنوعی وزن دنبه را از صفات وزن بدن پیش­بینی کرد. Manuscript profile -
Open Access Article
256 - Estimation of evaporation from Dez regulatory dam station pan using artificial neural network
mehdi najafvand derikvandi hossein eslamiMore rainfall in arid and semi-arid just evaporate into the atmosphere and so estimates the amount of water vapor in the water cycle will be important. Evaporation is dependent on various parameters and to its estimate needs for a different climate variables and the int MoreMore rainfall in arid and semi-arid just evaporate into the atmosphere and so estimates the amount of water vapor in the water cycle will be important. Evaporation is dependent on various parameters and to its estimate needs for a different climate variables and the interaction of these variables is very complex, so it must be accurate methods to be used in the evaporation study. In this study, artificial neural networks were used to estimate the pan evaporation of Dez regulating dam station. As ANN hyperbolic tangent function and the learning momentum was used. Multilayer Perceptron structure which used a network of six input neurons, three hidden layer and an output neuron was formed. Input layers include maximum temperature, minimum temperature, sunshine hours, average wind speed, relative humidity and an average rate of evaporation from water surface to the output layer. The relationship between climatic factors showed that the average temperature on the surface evaporation caused more than sunshine and wind speed. High coefficient of determination (92/0) between the actual data with simulated data with artificial neural network plus a small error (RMSE = 1.41) showed that the estimate accuracy is very high. Verification by t-test revealed no significant (P> 0.01) differences were between actual and estimated values. Manuscript profile -
Open Access Article
257 - Hargreaves Method Improves Accuracy in Estimating Reference Evapotranspiration Adjustment Weight With the Help of Artificial Neural Network and Decision Tree
omid mohtarami Mohammad Reza Hosseini Ruhollah Fattahi تیمور سهرابیOne of the most important components of the hydrological cycle is evapotranspiration which plays an important role in water resource management. In the present study the accuracy of evapotranspiration estimation by Hargreaves method and correction factor K was improved MoreOne of the most important components of the hydrological cycle is evapotranspiration which plays an important role in water resource management. In the present study the accuracy of evapotranspiration estimation by Hargreaves method and correction factor K was improved using the neural network and decision tree model M5. This coefficient is the ratio of Penman-Monteith evapotranspiration model is the method of Hargreaves. The data used in this study are the maximum and minimum temperatures and relative humidity in the period 2004-2013 Farokhshahr stations and airports in the region ShahrKord is cold and arid. Network Levenberg-Marquardt training algorithm is designed with a feedforward network and sigmoid tangent function is hidden in layers. Decision tree model was designed to help software WEKA. The results show that the neural network and decision tree model to model good performance, but the performance of the neural network model is more accurate correction factor. The results showed that the correction factor carefully before using the Hargreaves RMSE = 0.90 (Root Mean Square Error) Penman-Monteith than that this value after the correction factor to help RMSE = 0.69 and with the use of neural networks the correction factor to help decision tree to reach RMSE = 0.72. The results showed that after using a correction factor to the improved performance of Hargreaves. Manuscript profile -
Open Access Article
258 - Utilization of Artificial Neural Networks for Determining the Overflow Discharge of Marun Dam
Ebrahim Nohani valiolah partovi ziaFor more accurate measurement of the water flow, it has been always attempted to design structures with least errors and highest accuracy. Nowadays, the use of artificial neural networks (ANN) models has been rapidly grew mainly due to the fact that these models are not MoreFor more accurate measurement of the water flow, it has been always attempted to design structures with least errors and highest accuracy. Nowadays, the use of artificial neural networks (ANN) models has been rapidly grew mainly due to the fact that these models are not confined to the physical parameters. Artificial neural networks are based on use of embedded knowledge between input and output variables of a problem, regardless of physical aspects and these networks are able to extract inherent relation of the input and output and they can generalize the obtained relation to other situations and cases. In the present research, the information related to the overflow of Marun Storage Dam was adopted. The input parameters of ANN model are as follows: day, month, water surface elevation, water sharing percent and output parameters overflow discharge of storage dam. The models employed in artificial neural networks include FF, JEN, MLP and RBF. Moreover, the genetic algorithm (GA Manuscript profile -
Open Access Article
259 - Development of a Wavelet Hybrid Models for Estimating Regional Droughts in Siminehroud Basin
Erfan Rostam Zade alireza parvishiIn the present study, the drought of Siminehroud basin was investigated by intelligent Support Vector Machine (SVM) models, Artificial Neural Network (ANN) and Wavelet theory (W). Data from six rain gauge stations in the region were used and drought index was calculated MoreIn the present study, the drought of Siminehroud basin was investigated by intelligent Support Vector Machine (SVM) models, Artificial Neural Network (ANN) and Wavelet theory (W). Data from six rain gauge stations in the region were used and drought index was calculated in four time scales. The first-order autocorrelation was also selected as the optimal delay. Then the appropriate structure of the Artificial Neural Network was determined using Trial and Error Method and the three coefficients of the SVM model were determined and modeled. The results of evaluating individual models showed that there is no significant difference between two methods in predicting droughts. Then WANN and WSVM hybrid models were prepared. The results showed that the application of Wavelet theory greatly improved the performance of individual models and the amount of RMSE and MAE indices decreased by 19% and 21% and the correlation coefficient increased by 30%, respectively. Manuscript profile -
Open Access Article
260 - Forecasting the discharge of the Zayandeh Rood River at the Ghleeh Shahrokh station using deep learning techniques
Mohammad MehraniAbstract- Water discharge is a term in the water industry that refers to the amount of water that passes through a certain point per unit of time. Discharge rate is the amount of water that passes through a specific point such as a river,, water channel, dam valve, pipe MoreAbstract- Water discharge is a term in the water industry that refers to the amount of water that passes through a certain point per unit of time. Discharge rate is the amount of water that passes through a specific point such as a river,, water channel, dam valve, pipe or any other structure such as a faucet cartridge in a unit of time. In the metric system, water discharge rate is expressed in terms of cubic meters per second, cubic meters per hour, or liters per second. The unit of cubic meters per second is used for large flows such as rivers and large canals, and the unit of liters per second is used for the flow of water in wells and water that enters leaks. Measuring the discharge of the river has many effects on people's lives. Knowing the amount of water entering the areas of a river's catchment area is very important in agriculture, potential risks to human and animal life, industries, etc. Therefore, predicting river discharge can lead to effective management and prevent serious damage in the mentioned areas. According to the mentioned cases, the purpose of the presented paper is to predict the river discharge using deep learning techniques. In order to do this, the discharge of the Zayandeh Rood River at Qala Shahrokh station has been investigated and predicted using two techniques - ANFIS and LSTM. The simulation results show 93% to 94% accuracy in predicting the discharge of the studied river. Manuscript profile -
Open Access Article
261 - Development and Implementation of Principal Component Analysis Method for Monitoring of Gas Turbine
Samira Piri Niaragh Elham GhanbariGas turbines are complex and expensive machines that the cost of repairing unexpected failures is very high. There are many sensors installed in each gas turbine that record and collect large amounts of data. With the data mining of such big data, failure prediction is MoreGas turbines are complex and expensive machines that the cost of repairing unexpected failures is very high. There are many sensors installed in each gas turbine that record and collect large amounts of data. With the data mining of such big data, failure prediction is possible before the occurrence. The data set for the present study is the recorded quantities of sensors mounted on a 9-frame gas turbine in one of the country's power plants. The one column of data matrix rows was first labeled to identify healthy and defective row in each data sample. Then, by using the Principal Component Analysis method, the dimensions of the data matrix were reduced from seven to four dimensions and the main features were extracted. Following this, a model was developed by applying Artificial Neural Network method that was able to identify fault rows in the data matrix and identify the class of the data samples as healthy or defective. Accuracy, precision, and convergence of the model for two-to-six-dimensional model reductions were studied after machine learning was performed on 80% of the data. After matrix dimensionality reduction, and feature extraction by using "Principal Component Analysis" method, our well-designed model was also able to identify and classify the fault by using "Artificial Neural Network" method. In this thesis, it was found that our mode l by combining "Principal Component Analysis" method with "Artificial Neural Network" was able to show more than 90% precision with good accuracy and maximum degree of data matrix convergence. Moreover, it was able to specify the gas turbine fault class. Manuscript profile -
Open Access Article
262 - Assessment of Citizens' Satisfaction with Intra-City Transportation System in Hamedan City
Hamed AbbasiTransportation is one of the most important constituent parts of urban life that determines the shape and way of social and economic development of the city. In this regard, the quality of transportation in Hamadan city, which can be very useful in its development, was MoreTransportation is one of the most important constituent parts of urban life that determines the shape and way of social and economic development of the city. In this regard, the quality of transportation in Hamadan city, which can be very useful in its development, was considered. The nonparametric regression model and artificial neural network were used to assess the citizens' satisfaction from the transportation system. For this purpose, first, a questionnaire was developed based on three main indicators (equipment and facilities status, physical status, management status, and service delivery). Citizens' perspective was gathered. Then, using a nonparametric regression model, independent indicators and satisfaction as an independent variable were used. The correlation and root mean square of the output errors of this model were 0.914 and 0.334 respectively. In another approach, using an artificial neural network, a model with three intrinsic neuron structures, one hidden layer and one output neuron was constructed. The output correlation of this model was 0.998 and the mean square error of the error was about 6 times lower than regression model. The results showed that the neural network model with both linear and nonlinear correlation estimates is more versatile and more suitable than nonparametric regression. On the other hand, price indices with coefficient (0.885), equality and welfare with (0.795) and decrease in demand for travel (0.790) are the most effective indicators for citizens' satisfaction with urban transport network. Manuscript profile -
Open Access Article
263 - Strategic planning of Piranshahr City
leili ebrahimi Mohammad Rahim Ezzatolah MafiAwareness of the strengths and weaknesses of cities is essential for present plans, policies and plans of the city. In fact, the use of economic, social, physical, etc. indicators can be a suitable Criterion for determining the status of cities, and fields of solving th MoreAwareness of the strengths and weaknesses of cities is essential for present plans, policies and plans of the city. In fact, the use of economic, social, physical, etc. indicators can be a suitable Criterion for determining the status of cities, and fields of solving the problems and failures are created to achieve the desired and actual development. Accordingly, the purpose of this paper is to determine the position of Piranshahr city in terms of strategic planning indices and prioritizing the city development strategies. The research is descriptive and analytical. The sample population in this study consisted of three groups: urban managers, urban elites and citizens the sample size was identified 382 people for the citizens group according to the Cochran formula. In the group of managers 50 people and in the elite group 50 were selected as samples. Positioning of city was by using SWOT model and Prioritization of city developing strategies was performed by the SWOT model and prioritizing the city development strategies using a combination of Multi Criteria Decision-Making method and Kohonen neural network in MATLAB and EXCEL softwares. The results showed that the city of Piranshahr is in a competitive position strategically. Also the output of Kohonen neural network identified strategy of developing of foreign exchanges and strengthening of Tamarchin boundry market in order to shaping a sustainable economy based on internal - foreign trade as the best city development strategy. Manuscript profile -
Open Access Article
264 - Application of artificial neural network and Cellular AutomataIn modeling and predicting land use changes in Bavanat city
Marzieh MogholiIntroduction: Today, due to the high value of land and the limitation of natural resources in the city of Bowanat, it is very important to predict land use changes in this city.Research Aime: determining the level of ability in modeling the localization phenomena in the MoreIntroduction: Today, due to the high value of land and the limitation of natural resources in the city of Bowanat, it is very important to predict land use changes in this city.Research Aime: determining the level of ability in modeling the localization phenomena in the city of Bowanat is one of the main goals of the research. Methodology: considering the practicality and development in this research of artificial neural networks for calibrating the model for the effective factors in the city. Bowanat has been used and ENVI and Arc GIS image processing software have been used.Methodology: Due to practicality and development in this research, artificial neural networks were used to calibrate the model for effective factors in the city of Bowanat, and ENVI and Arc GIS image processing software were used.Studied Areas: Bowanat city is located 240 km from Shiraz city with an area of 4992.2 square kilometers, which is located at 30.46 degrees north and 53.67 degrees east.Results: In the design of urban growth modeling in Bowanat between 2003 and 2018 using artificial neural network, it was observed that for two main reasons, the mentioned model is suitable for predicting land use changes in Bowanat city, the first reason being the ability of the CA model and the reason The second is to achieve a model for urban change and expansion by changing urban land use.Conclusion: After examining the findings, it was found that the road network is one of the most important factors in the growth and expansion of Bowanat city, and in addition, the percentage of land slope is one of the effective parameters in the modeling of Bowanat city.Keywords: Land use, Fuzzy Logic, Artificial neural network, Bavanat city. Manuscript profile -
Open Access Article
265 - Application of artificial neural networks in modeling urban physical development (Case study: Rasht city)
tala abedi golamreza miri parviz rezaei reza zareiIntroduction: The physical development of cities is increasing day by day. Correct management of this development from various aspects is among the important issues that must be considered. There are many methods for predicting and determining the direction of urban dev MoreIntroduction: The physical development of cities is increasing day by day. Correct management of this development from various aspects is among the important issues that must be considered. There are many methods for predicting and determining the direction of urban development, one of these methods for determining suitable areas is the method based on neural networks.The purpose of the research: The purpose of this research is to model the development of the city of Rasht in the last 20 years and predict the directions of development of this city until 2032.Research methodology :By using ETM+ Landsat 7 and 8 satellite images of 2002, 2012 and 2021 of Rasht city and with GIS software, images with suitable band composition are prepared and then the images are classified using Multi Layer Perceptron Artificial Neural Network (MLP) method. The indicators considered for the neighborhood model of urban areas are the distance from urban points, the distance to the central areas of the city, and the distance to the main streets and roads.The geographical scope of the research:Rasht city, the capital of Gilan province, is located at 49 degrees 35 minutes 45 seconds east longitude and 37 degrees 16 minutes 30 seconds north latitude from the Greenwich meridian, and its area is about 10,240 hectares.Findings and discussion:In this model, in the training mode of the first stage (input, applying 4 indexes on the images of 2002), the network performed 104 iterations, and the lowest error rate, which is evaluated by the crossentropy criterion, was equal to 0.058526 in the 98th iteration. In the second step, the input of the model was to apply 4 indicators on the images of 2012, and the lowest error rate was evaluated as 0.076657.Results :In total, the model has been able to predict the development of Rasht city in 2012, 95.9% and for 2021, 93.8%, which can be acceptable. The model error in this first part was 1.4% and in the second part was 2.6%. By examining the 20-year period of physical development, the development directions of Rasht city in 2032 were predicted. Manuscript profile -
Open Access Article
266 - Expectation of Chamomile Fundamental Oil Abdicate by Using the Artificial Neural Network System
Nazanin Khakipour Mahtab PayandehThe aim of this research was to forecast the proportion and production of chamomile essential oils by employing an artificial neural network system reliant on specific soil physicochemical characteristics. Various chamomile cultivation sites were explored, and 100 soil MoreThe aim of this research was to forecast the proportion and production of chamomile essential oils by employing an artificial neural network system reliant on specific soil physicochemical characteristics. Various chamomile cultivation sites were explored, and 100 soil samples were transported to the greenhouse. The pH, EC, K, OM (organic matter), CCE (calcium carbonate equivalent), and clay content in the soils ranged from 8.75 to 7.94, 1.6 to 1.0, 381 to 135, 2.30 to 0.22, 69 to 16, and 55.6 to 32.0, respectively. Growth parameters, essential oil percentage, and yield were measured. The artificial neural network modeling aimed to predict essential oil concentration and yield using three sets of soil properties as predictors: Nitrogen (N), phosphorus (P), potassium (K), and clay; pH, EC, organic matter (OM), and clay; CCE, clay, silt, sand, N, P, K, OM, pH, and EC. Consequently, three pedotransfer functions (PTFs) were formulated using the multi-layer perceptron (MLP) with the Levenberg-Marquardt training algorithm to estimate chamomile essential oil content. The evaluation of results indicated that the third PTF (PTF3), developed using all independent variables, exhibited the highest accuracy and reliability. Furthermore, the findings suggested the feasibility of predicting chamomile essential oil concentration and yield based on soil physicochemical properties. This has significant implications for land suitability assessments, identifying areas conducive to chamomile cultivation, and planning for essential oil yields. Manuscript profile -
Open Access Article
267 - شناسایی گیاهان آپارتمانی بر اساس ویژگی های تصویری با کمک شبکه عصبی
نرگس قانعی قوشخانه عباس روحانی محمودرضا گلزاریان فاطمه کاظمیدر این مقاله سامانه بینایی ماشینی مبتنی بر شبکه عصبی برای شناسایی 12 گیاه آپارتمانی توسعه داده شد. از سامانه پردازش تصویر برای استخراج 41 ویژگی رنگی، بافتی و شکلی از تصاویر رو و پشت برگ گیاه استفاده گردید. ویژگی­های استخراج یافته به عنوان معیار تشخیص و و Moreدر این مقاله سامانه بینایی ماشینی مبتنی بر شبکه عصبی برای شناسایی 12 گیاه آپارتمانی توسعه داده شد. از سامانه پردازش تصویر برای استخراج 41 ویژگی رنگی، بافتی و شکلی از تصاویر رو و پشت برگ گیاه استفاده گردید. ویژگی­های استخراج یافته به عنوان معیار تشخیص و ورودی به شبکه عصبی داده شد. شبکه عصبی پرسپترون چند لایه (MLP) با الگوریتم آموزش، الگوریتم فاکتور کاهش نرخ یادگیری (BDLRF) به عنوان طبقه­بندی کننده استفاده گردید. طبقه­بندی در سه مرحله براساس قابلیت و قدرت ویژگیها در شناسایی گیاهان انجام شد. معیار قابلیت داشتن در هر مرحله با استفاده از قدرت تفکیک پذیری کلاسی گیاهان بررسی گردید. در این روش طبقه­بندی، هر مرحله نیاز به تعداد کمی از ویژگیها دارد؛ در نتیجه سرعت و دقت آن میتواند بسیار بالا باشد. نتایج نشان داد که دقت طبقه­بندی گیاهان در سه مرحله به 100% میرسد. همچنین ویژگیهای بهینه برای طبقه­بندی شامل سه مرحلهی ورودی از ویژگیهای موفولوژیکی (شکلی)، ویژگیهای رنگی HSI استخراج یافته از پشت برگ و ویژگیهای بافتی HSI استخراج یافته از پشت برگها میشود. Manuscript profile -
Open Access Article
268 - ارائه یک مدل شبکه عصبی RBF برای پیشبینی روزهای کاری عملیات خاکورزی تولیدات محصولات
ارمغان کوثری مقدم عباس روحانی Lobat Kosari-Moghaddam مهدی اسماعیل پور تروجنیهدف از این مطالعه تعیین احتمال روزهای کاری (PWD) برای عملیات خاکورزی با استفاده از دادههای هواشناسی و به کمک روش مدلسازی رگرسیونی خطی چندگانه (MLR) و شبکه عصبی مصنوعی توابع پایهای شعاعی (RBF) بوده است. در هر دو مدل، هفت متغیر شامل دمای متوسط، بیشینه و کمینه، رطوبت نس Moreهدف از این مطالعه تعیین احتمال روزهای کاری (PWD) برای عملیات خاکورزی با استفاده از دادههای هواشناسی و به کمک روش مدلسازی رگرسیونی خطی چندگانه (MLR) و شبکه عصبی مصنوعی توابع پایهای شعاعی (RBF) بوده است. در هر دو مدل، هفت متغیر شامل دمای متوسط، بیشینه و کمینه، رطوبت نسبی، بارندگی، سرعت باد و تبخیر و تعرق بر پایه روزانه به عنوان پارامترهای ورودی در نظر گرفته شدند. احتمال روزهای کاری نیز به عنوان خروجی مدلهای ایجاد شده، در نظر گرفته شد. معیارهای عملکردی شامل مجذور مربعات خطا (RMSE)، میانگین درصد خطای مطلق (MAPE) و ضریب تبیین (R2) بودند. نتایج نشان داد که مقادیر R2 برای مدلهای MLR و RBF به ترتیب برابر 78/0 و 99/0 بوده است. هر دو مدل عملکرد قابل قبولی داشتند؛ اما مدل RBF عملکرد دقیقتری نسبت به مدل MLR داشت. مقادیر RMSE و MAPE نیز برای مدل RBF کمتر از مدل MLR بدست آمد. بنابراین مدل RBF به عنوان بهترین مدل برای پیشبینی احتمال روزهای کاری انتخاب شد. علاوه بر این، نتایج این مدلها با یک مدل رطوبت خاک که پیشتر ارائه شده بود، مقایسه شد. نشان داده شد که نتایج مدلهای مورد مطالعه با نتایج مدل رطوبت خاک سازگاری خوبی داشته است، اگرچه که مدل RBF بالاترین ضریب تبیین را داشت (R2=99%). در نهایت میتوان بیان نمود که مدل RBF ایجاد شده میتواند برای پیشبینی احتمال روزهای کاری در راستای سیاستگذاریهای مدیریتی در بخش کشاورزی مورد استفاده قرار گیرد. Manuscript profile -
Open Access Article
269 - پیشبینی صادرات زعفران ایران با مقایسه الگوریتم های یادگیری ماشین
علیرضا امیرتیموری منصور صوفی مهدی همایونفر مهدی فداییواردات و صادرات در همه کشورها نقش مهمی در رشد اقتصادی ایفا میکنند. بنابراین، انتخاب محصولات مناسب، باعث افزایش رقابتپذیری کشور در تجارت جهانی میشود. زعفران یکی از مهمترین و متمایزترین محصولات غیرنفتی ایران برای صادرات است. هدف این مطالعه، پیشبینی صادرات زعفران از ط Moreواردات و صادرات در همه کشورها نقش مهمی در رشد اقتصادی ایفا میکنند. بنابراین، انتخاب محصولات مناسب، باعث افزایش رقابتپذیری کشور در تجارت جهانی میشود. زعفران یکی از مهمترین و متمایزترین محصولات غیرنفتی ایران برای صادرات است. هدف این مطالعه، پیشبینی صادرات زعفران از طریق سه الگوریتم دادهکاوی و انتخاب الگوریتم مناسب در پیشبینی است. دوره نمونه مدلهای پیشبینی شامل دادههای صادرات زعفران ایران از سال ۲۰۱۲ تا ۲۰۱۹ است که از انجمن زعفران ایران جمعآوری شدهاند. پس از انجام مراحل آمادهسازی داده، پیشبینی صادرات زعفران با استفاده از سه الگوریتم دادهکاوی شامل شبکه عصبی مصنوعی، یادگیری عمیق و درخت تقویت گرادیانی انجام شد. برای انتخاب یک مدل پیشبینی بهتر، اعتبار مدل نقش مهمی ایفا میکند. صحت پیشبینی سه مدل طراحی شده به کمک خطای مطلق ( 036/0 = شبکهی عصبی مصنوعی، 031/0 = یادگیری عمیق شبکه، 047/0 = درخت تقویت گرادیانی)، معیار R2 (045/0 = شبکهی عصبی مصنوعی، 044/0 = یادگیری عمیق شبکه، 073/0 = درخت تقویت گرادیانی) و همبستگی (95/0 = شبکهی عصبی مصنوعی، 98/0 = یادگیری عمیق شبکه، 97/0 = درخت تقویت گرادیانی) اندازهگیری شدند. براساس یافتهها، تمامی این سه مدل طراحی شده دقیق هستند و خطای پیشبینی آنها بسیار کم و نزدیک به هم است. اما با تفاوت ناچیز، شبکه یادگیری عمیق کمترین خطا را دارد. نتایج میتوانند برای برنامهریزی دقیقتر صادرات زعفران مفید باشند. Manuscript profile -
Open Access Article
270 - تخمین مقاومت فشاری تک محوری و مدول الاستیسیته نمونه های کنگلومرا با استفاده از رگرسیون و شبکه عصبی مصنوعی
مجتبی حیدری بهروز رفیعی مهران نوری غلامرضا خانلری علی اکبر مومنی -
Open Access Article
271 - پیش بینی تراز آب زیرزمینی با استفاده از مدل شبکه های عصبی مصنوعی (مطالعه موردی: دشت شبستر)
زینب مختاری امیرحسین ناظمی عطااله ندیری -
Open Access Article
272 - Reviewing the websites of Tehran Municipality and providing appropriate data mining solutions
shaysteh shojaei karizaki sudabeh Shapoori hajar zareiObjective: The main purpose of this study is to identify and analyze different types of data on the website of Tehran Municipality and to provide appropriate data mining solutions. Method: This research is fundamental and in terms of nature it can be considered analytic MoreObjective: The main purpose of this study is to identify and analyze different types of data on the website of Tehran Municipality and to provide appropriate data mining solutions. Method: This research is fundamental and in terms of nature it can be considered analytical. The data collection method was field and the statistical population of 47 sites were selected among 220 domains of Tehran Municipality and data mining techniques were used for analysis and the source of data collection is web analytics and tools used by Google Analytics. Results: The accuracy of the normal neural network algorithm is equal to 99.25% and the RMS standard of the normal neural network algorithm is equal to 0.159. The accuracy of the decision tree algorithm is 99.80% and the MSI criterion of the decision tree algorithm is 0.003 and finally the RMS criterion of the decision tree algorithm is 0.045. The accuracy of the CNN algorithm is equal to 99.81% and finally the RMS criterion of the CNN algorithm is equal to 0.035. Conclusion: Based on the obtained findings, the DB Scan method is equal to other basic methods for analyzing data of Tehran Municipality websites and has a higher accuracy than other methods. Manuscript profile -
Open Access Article
273 - بررسی معیارهای متفاوت برای منظم کردن اجزاهای اصلی به منظور ایجاد یک مدل QSPR برای پیش بینی نقطه های ذوب
ولی زارع شاه­ آبادی فاطمه عباسی­ تبار -
Open Access Article
274 - کاربرد شبکه عصبی مصنوعی در پیشگویی بازده استخراج یونهای روی از نمونههای حقیقی با استفاده از مولکول نگاری بسپاری
سید حسین هاشمی مسعود کیخوایی مجید میرمقدم محمد شاکری -
Open Access Article
275 - مدلسازی فرایند تبدیل خشک متان بهکمک پلاسما با استفاده از شبکه عصبی مصنوعی و الگوریتم ژنتیک
سیدمرتضی فاضلی فاطمه راوری حمیدرضا بزرگ زاده جعفر صادق زاده اهری -
Open Access Article
276 - روشهای خطی و غیرخطی ارتباط کمیساختار- فعالیت جهت پیشبینی فعالیت دارویی برخی از مشتقات آمینواسیدها
مهدی نکویی مجید محمدحسینی مهدی رحیمی عبدالرضا علویقرهباغ -
Open Access Article
277 - Presenting a model for predicting the price of digital currency in conditions of environmental uncertainty with a fuzzy artificial neural network
mohammad hasan darvish motevali shirin aminiAbstarctIn this research, using the method of fuzzy neural networks, the price of Bitcoin is predicted. In order to identify the appropriate criteria in this research in order to predict the price of Bitcoin, we have used previous studies and researches in this field in MoreAbstarctIn this research, using the method of fuzzy neural networks, the price of Bitcoin is predicted. In order to identify the appropriate criteria in this research in order to predict the price of Bitcoin, we have used previous studies and researches in this field in the first stage. In the following, using interviews with experts and experts in this field, the available information about Bitcoin became the final factors. Research information was collected using related sites and identified criteria. In this way, we first normalized the collected data. In the next step, by entering the normalized information into the MATLAB software and using the designed toolbox and using the fuzzy neural network method, Bitcoin price was predicted. In this way, 60% of the input data, which includes 1330 data, was considered as training data and 40% of the data, which is 887 data, was considered as testing. The research results show high accuracy prediction using the proposed method. As the error was considered in two cases, a small value was calculated for the error of the method. Keywords: prediction, bitcoin price, fuzzy neural network. Manuscript profile -
Open Access Article
278 - Predicting the Factors Affecting the Risk of Future Stock Price Decline Based on Circuit Radius Neural Network
Reza Ataeizadeh Roya DarabiThe main objective of this study is to predict the factors affecting the risk of future stock price decline under the Circuit Radius Neural Network Method in companies admitted to the Tehran Stock Exchange during the period from 2010 to 2015. To collect theoretical foun MoreThe main objective of this study is to predict the factors affecting the risk of future stock price decline under the Circuit Radius Neural Network Method in companies admitted to the Tehran Stock Exchange during the period from 2010 to 2015. To collect theoretical foundations of the research, library method, books, Dissertations and articles have been used. To collect statistical information financial statements and accompanying notes have been utilized. For the analysis of this study Circuit Radius Neural System and SPSS23 software have been used. The results of the research indicate that predicting the factors affecting the risk of future stock price decline is possible on the basis of the Circuit Radius Neural Network Method. The research findings also show that the unusual optional cost, the opportunity to grow, the abnormal operating cash flows, the capital structure, and the size of the company affect the risk of future stock price decline respectively as the first, second, third, fourth and fifth priority. Manuscript profile -
Open Access Article
279 - بکارگیری شبکههای عصبی Fuzzy ARTMAP برای تشخیص صعود ناگهانی سیگنالهای EEG با استفاده از استخراج ویژگی توسط موجک
فاطمه صفری علی فرخی نعمت طالبی این مقاله در پی معرفی دو سیستم کلاسبندی مبتنی بر شبکههای عصبی Fuzzy Artmap برای تشخیص اتوماتیک حدودهای ناگهانی در امواج الکترد آلفا نوگرافی (EEG) 19 کانال اشخاص میباشد. این الگوریتم سریع و نتایح قابل قبولی عرضه مینمایند. سیگنالهای EEG به چهار زیر باند با استف More این مقاله در پی معرفی دو سیستم کلاسبندی مبتنی بر شبکههای عصبی Fuzzy Artmap برای تشخیص اتوماتیک حدودهای ناگهانی در امواج الکترد آلفا نوگرافی (EEG) 19 کانال اشخاص میباشد. این الگوریتم سریع و نتایح قابل قبولی عرضه مینمایند. سیگنالهای EEG به چهار زیر باند با استفاده از تبدیل ویولت گسسته تقسیم بندی شدهاند. وروردیهای شبکه شامل دو ویژگی متفاوت هستند که از زیرباندهای 3 و 4 استخراج میشوند. عملکرد این کلاسبندی کنندهها در این مقاله معرفی شده و باهم و دیگر سیستمها مشابه مطابق با مقادیر حساسیت، ویژگی و انتخاب پذیری مقایسه گشتهاند. Manuscript profile -
Open Access Article
280 - .Application of Meta-Heuristic Algorithms in Predicting Financial Distress using intra-corporate (Financial and non-financial) and Economic Variables (Grasshopper Optimization and Ant Colony Algorithms)
فریدون مرادی احمد یعقوب نژاد آزیتا جهانشاد. Abstract The purpose of this study is investigating the capability of Grasshopper Optimization Algorithm (GOA) in more accurately predicting the financial distress by-using intra-corporate (financial and non-financial) and economic variables. The method of this rese More. Abstract The purpose of this study is investigating the capability of Grasshopper Optimization Algorithm (GOA) in more accurately predicting the financial distress by-using intra-corporate (financial and non-financial) and economic variables. The method of this research is improving the performance of the basic model of Multilayer Perceptron Artificial Neural Network (ANN-MLP) by-using a hybrid model with GOA (MLP-GOA) and Ant Colony Optimization Algorithm (MLP-ACO). The statistical research population of companies active in Tehran Stock Exchange during a 7-year period (from 1391 to 1397) included 476 companies, and finally, after systematic elimination, there were 289 qualified companies (including 2023 observation year-company). Checked and screened. The results showed the ability of ANN-MLP model to predict financial distress by-using financial and non-financial variables, and in addition the hybrid models (MLP-GOA and MLP-ACO) had been improved this ability. The accuracy of the MLP-GOA model for the year t, year t-1and year t-2 (before financial distress occurs), respectively are 97.30%, 94.53% and 91.30% that higher than the accuracy of the basic model and the hybrid MLP-ACO model. Although, entering the economic variables has increased the capability of all models significantly but the results showed that the financial distress is more affected by intra-corporate variables and the effect of economic variables has already been considered through the effect on financial events recorded in the accounting system. The results of this study can be used by company managers, banks and rating and credit institutions, insurance companies, financial analysts, investors and investment companies in assessing the risk of financial distress to make appropriate decisions and actions. Manuscript profile -
Open Access Article
281 - پیشبینی شاخص سهام با استفاده از ترکیب شبکه عصبی مصنوعی و مدلهای فرا ابتکاری جستجوی هارمونی و الگوریتم ژنتیک
مریم دولو تکتم حیدری -
Open Access Article
282 - بررسی اثر روابط اعطای تسهیلات بر هزینه مبادله تسهیلات بانکی: مطالعه موردی بانک کشاورزی
عباس عرب مازار مهرداد نعمتی امیر درویشی -
Open Access Article
283 - Identifying and Sensitivity Analysis the Effective Factors on Main Forming Groups of Iran’s Inflation: The Artificial Neural Network Approach
seyed sajad alam al hodaa samaneh tarighi mehdi shaban zadeh amin kajoeipoorAbstract Since the inflation has numerous effects on various aspects of economic and social, many economists believe the importance of inflation is higher than other economic indicators.With this approach, in this study the effective factors on main forming groups of I MoreAbstract Since the inflation has numerous effects on various aspects of economic and social, many economists believe the importance of inflation is higher than other economic indicators.With this approach, in this study the effective factors on main forming groups of Iran’s inflation in the Fourth Development Plan identified and have been examined. For achieve this target, among the twelve main forming groups of consumer price index (CPI), the important forming groups of CPI have been identified using Neural Network Sensitivity Analysis. Then by definition the causes and origins of inflation in Iran using various economic theories and domestic studies, the effects of these factors has been studied on the main commodity groups of CPI. The result of this study showed, among the main groups of commodities, Food and beverage group (Group 1), Housing , water, electricity, gas and other fuels Group (Group 4) and transport group (Group 7) were more important Compared to the other groups. So that, these groups explained 15.22%, 13.19% and 12.23% of the total change in the CPI during the period under review, respectively. The results of this study indicated the commodities of group 1 have been most affected by changes in liquidity, exchange rates and the GDP gap. Also commodities of group 4 and 7 have been most impressed by the rate of return on rent housing in urban areas and liquidity, inflation expectations and exchange rate, respectively Manuscript profile -
Open Access Article
284 - Predictability Test of Stock Market Price Index in Iran Investment Market and comparing Linear and Nonlinear models predictability potentials
Karim Emami Ghodratollah EmamverdiSince the highly complicated Time Series such as Stock Market Prices are usually stochastic, their changes are assumed to be unpredictable. Some tests which have been used to study the statistical observations related to the economical variables e.g. Stock Market Price, MoreSince the highly complicated Time Series such as Stock Market Prices are usually stochastic, their changes are assumed to be unpredictable. Some tests which have been used to study the statistical observations related to the economical variables e.g. Stock Market Price, are often go wrong while encountering the chaotic data and recognize them as stochastic ones, though these data are actually generated from the deterministic systems which bear few tribulations. For this reason the predictable and non-linear tests such as HURST, BDS, Runs Test, and Correlation Dimension have been used to study the existence of deterministic chaotic trend and non-linear process in Time Series of Daily Stock Market Price Index of TEHRAN STOCK EXCHANGE from 23 rd October, 2000 to 24 th September, 2002. The result of the above mentioned tests shows the predictability and the existence of a non-linear process in the sample data. After the illustration of predictability and the non-linear process in daily stock index data, then the linear time series models (AR), non-linear (GARCH) and Artificial Neural Network (ANN) have been estimated to present a suitable model for predicting the Stock Price Index. Comparing the potential of predictability of these models by such criteria as: CDC, RMSE, MAE, MAPE and U-THEIL inequality coefficient, it has been revealed that there is the highest potential of predictability in Artificial Neural Network models than the other ones Manuscript profile -
Open Access Article
285 - Presenting the Forecasting Model of Bitcoin Return Using the hybrid Method of Deep Learning - Signal Decomposition Algorithm (CEEMD-DL)
sakineh sayyadi nezhad Ali Esmaeil Zadeh Mohammad Reza RostamiAbstract With the increasing popularity and widespread use of cryptocurrencies, the creation and development of methods for predicting price movements in this field has attracted a lot of attention. In between, recent developments in deep learning (DL) models with stru MoreAbstract With the increasing popularity and widespread use of cryptocurrencies, the creation and development of methods for predicting price movements in this field has attracted a lot of attention. In between, recent developments in deep learning (DL) models with structures such as long-short-term memory (LSTM) and convolutional neural network (CNN) have made improvements in the analysis of this type of data. Another approach that can be effective in the analysis of cryptocurrencies time series is the decomposition through algorithms such as complete integrated empirical mode decomposition (CEEMD). Considering the importance of forecasting in the cryptocurrencies field, in this research, by combining deep learning models and complete integrated empirical mode decomposition (CEEMD), The hybrid CEEMD-DL(LSTM) model has been used to forecast the bitcoin return (as the most popular currency). In this regard, the daily data of the total index of the Tehran Stock Exchange was used in the period of 2013/01/01 – 2022/05/28 and the results obtained were compared with the results of competing models based on efficiency measurement criteria. Based on the obtained results, the use of the introduced model (CEEMD-DL(LSTM)) has increased the efficiency and accuracy of bitcoin return forecasting. Accordingly, the use of this model in this field is suggested. Manuscript profile -
Open Access Article
286 - مقایسه برآورد تلاطم بازارهای مالی با استفاده از مدل رگرسیون و مدل شبکه عصبی
محمد عظیم خدایاری احمد یعقوب نژاد مریم خلیلی عراقی -
Open Access Article
287 - Presenting a Comprehensive Model for Measuring the Liquidity Risk of Banks Listed on the Tehran Stock Exchange (Case Study: Mellat Bank)
Toraj Azari Mojtaba Tastori Reza Tehrani AbstractLack of liquidity management of banks is one of the most important risks for any bank and lack of attention to liquidity risk leads to irreparable consequences. Preventing liquidity risk requires a comprehensive measurement method but liquidity risk is com More AbstractLack of liquidity management of banks is one of the most important risks for any bank and lack of attention to liquidity risk leads to irreparable consequences. Preventing liquidity risk requires a comprehensive measurement method but liquidity risk is complicated issue, and this complexity makes it difficult to provide a proper definition. In addition, defining liquidity risk determinants and formulation of the related objective function to measurement its value is a difficult task. To address these problems and assess liquidity risk and its key factors, in this study we propose a model that uses artificial neural networks and Bayesian networks. Design and implementation of this model includes several algorithms and experiments to validate the model. In this paper, we have used Levenberg-Marquardt and Genetic optimization algorithms to teach artificial neural networks. We have also implemented a case study in Bank Mellat to demonstrate the feasibility, efficiency, accuracy and flexibility of the research liquidity risk measurement model. Manuscript profile -
Open Access Article
288 - Analysis of knowledge-based organizational culture components using artificial neural network
Somayeh hosseinzadeh Toraj Mojibi Sayyedmahdi alvani javad rezaiyanThis study analyzes the components of knowledge-based organizational culture in order to achieve performance effectiveness using artificial neural network.The present study, from an application point of view, is Applied Research, which was applied using a mixed Explorat MoreThis study analyzes the components of knowledge-based organizational culture in order to achieve performance effectiveness using artificial neural network.The present study, from an application point of view, is Applied Research, which was applied using a mixed Exploratory Research type method. To prepare the research literature, the method of studying documents and their results was applied and also intensive face-to-face interviews with 20 university experts were accomplished. After recording the interviews, the data were analyzed qualitatively using a context analysis method. The research population, in the quantitative part, consisted of 50 administrative experts of related field who were selected using objective sampling. To measure the effectiveness of each the components of knowledge-based organizational culture and analysis of their sensitivity a Multilayer feedforward neural networks with back propagation model was used.Findings showed that the sensitivity of performance effectiveness to knowledge-based organizational culture is 4.39%. Also, knowledge-based organizational culture was identified with five components, and among its components, the two components of knowledge-based leadership and trust culture were identified as the components to which performance effectiveness is most sensitive. Manuscript profile -
Open Access Article
289 - Designing Credit Risk Early-warning System for Individual and Corporate Customers of the Banks using Neural Network Models, Survival Probability Function and Support Vector Machine
Roya Derakhshani Mirfeiz Fallah hosein jahangirnia Reza Gholami jamkarani Hamidreza kordlouieCredit risk is the probability of default of the borrower or the counterparty of the bank in fulfilling its obligations, according to the agreed terms. In other words, uncertainty about receiving future investment income is called risk, which is of great importance in b MoreCredit risk is the probability of default of the borrower or the counterparty of the bank in fulfilling its obligations, according to the agreed terms. In other words, uncertainty about receiving future investment income is called risk, which is of great importance in banks. The purpose of this article is to estimate the credit risk of individual and corporate customers. In this study, the statistical information of 400 individual customers and7500 corporate customers was used. In this regard, the results of neural network model and support vector machine model have been compared. The obtained results have shown that the components considered in this study based on their personal, financial and economic characteristics had significant effects on the probability of customer default and credit risk calculation. Also, the results of this study showed that the application of control policies at the beginning of the repayment period suggests facilities that have the highest probability of default with long life and high repayment. The comparison of the results of the prediction accuracy shows the higher explanatory power of the support vector machine model and the use of the survival probability function than the simple neural network model for both groups of customers. Manuscript profile -
Open Access Article
290 - ارزیابی مولفه های BIS با استفاده از منطق فازی و شبکه عصبی
فرشته اسدالهیان حامد شاکریان -
Open Access Article
291 - پیش بینی مدیریت سود با استفاده از شبکه عصبی و درخت تصمیم
مهدی صالحی لاله فرخی پیله رود -
Open Access Article
292 - EPS forecast modeling using neural networks - Fuzzy
A.A Anvari عادل Azar محمد NoroziEarnings per share prediction and its changes as an economic events, past, were interested for investors, managers, financial analysts and creditors. This interest is because of the use of earnings in share valuation models, improving efficient performing of capital mar MoreEarnings per share prediction and its changes as an economic events, past, were interested for investors, managers, financial analysts and creditors. This interest is because of the use of earnings in share valuation models, improving efficient performing of capital markets, and evaluating solvency and evaluating of firm performance. The purpose of this paper is to earnings per share prediction using neural-fuzzy networks, MLP, GMDH, and determine most preferable model using four measures of evaluating performance. So, companies listed in TSE was chosen as statistical population and statistical sample is consisted of 500 firm-year 24 active industry 1386 to 1390 were chosen randomly using clustering sampling. The results show that neural-fuzzy networks is the most preferable comparing with neural networks, MLP, and GDMH, in all of four measures of evaluating performance, that it is showing of high power of this kind of networks in identifying dominant patterns of data and existence of non-liner relations of some accounting variables with EPS. So, the accuracy of neural-fuzzy networks predictions is more than MLP and GDMH, and is more suitable for EPS prediction. Manuscript profile -
Open Access Article
293 - Barriers in determining the stock price by ANN
رویا Darabi ربابه KarimiThe Extant Preventives in Determining Price of Shaves by Using Artificial Neural Network(In Metal and Mineral Industries) Roya Darabi Robabeh Karimi Raste Kenari (Received: 10/Apr/2014; Accepted: 12/Jun/2014) Abstract This study aims to investigate the extant preventive MoreThe Extant Preventives in Determining Price of Shaves by Using Artificial Neural Network(In Metal and Mineral Industries) Roya Darabi Robabeh Karimi Raste Kenari (Received: 10/Apr/2014; Accepted: 12/Jun/2014) Abstract This study aims to investigate the extant preventives in determining price of shaves by using artificial nerve network in metal and mineral industries companies accepted in Tehran Securities Exchange. We applied two statistical analysis and nerve network methods for examination of hypnotizes. A questionnaire was created for statistical analysis method and statistical society including senior experts of securities exchange course and instructors Tehran Azad University, who are familiar with the concepts of nerve network and also forecasting shares prices. The research hypnotizes were dealt with use of t test and score, ultimately all hypnotizes were approved. In the nerve network, the assumptions of research were studied by the use of nerve network and after the distribution of the error and Market-Loren berg educational model and it was determined that while all the indexes were entered into the network as input, the for forecasts of prices do not enter into the network as precisely as the indexes do enter. Also, the rate of network error was increasing. The results of nerve network corresponded with the results of statistical analysis. In other words, in both methods, the indexes have been marked as obstacles for forecasting the shares prices by the use of nerve network method. Key Words: Nerve Network, Forecast of Shares Price, Relative Strength Index, Rate of Shares Cost Index, After Distribution of Error. Manuscript profile -
Open Access Article
294 - The Comparison of Financial Crisis Prediction Strength of Different Artificial Intelligence Techniques
Zahra Pourzamani Hassan kalantariRapid technological advances and vast environmental changes, leading to increasing competition and limit access to benefits and likely to suffer financial crisis has increased. Purpose of this study is investigating financial crisis prediction strength of different arti MoreRapid technological advances and vast environmental changes, leading to increasing competition and limit access to benefits and likely to suffer financial crisis has increased. Purpose of this study is investigating financial crisis prediction strength of different artificial intelligence techniques(linear and nonlinear genetic algorithm and neural network). Based on available information and statistics, of all companies listed in Tehran Stock Exchange, 72 companies have been subject to Article 141 trade law and 72 companies have not been subject to this Article was elected. Results of Mc-Nemar test for genetic algorithms techniques and neural network showed that there are not significant differences between linear and nonlinear genetic algorithms with neural network. Although the predictive accuracy of nonlinear genetic algorithm(90%) and linear genetic algorithms(80%) is more than of the neural network(70%) but this difference is not statistically significant. Manuscript profile -
Open Access Article
295 - The impact on the quality of earnings bankruptcy prediction using artificial neural network
بیتا Mashikhi H.R GanjiThe Effect of Earnings Quality on Predicting Bankruptcy by Using Artificial Neural Networks Bita Mashayekhi Hamidreza Ganji (Received: 16/Apr/2014; Accepted: 19/Jun/2014) Abstract Predicting of entities’ going- concern assumption in the future periods is an import MoreThe Effect of Earnings Quality on Predicting Bankruptcy by Using Artificial Neural Networks Bita Mashayekhi Hamidreza Ganji (Received: 16/Apr/2014; Accepted: 19/Jun/2014) Abstract Predicting of entities’ going- concern assumption in the future periods is an important element in decision-making process of many investors. So, ing the predictor variables have been discussed as a challenging issue in the literature of bankruptcy prediction that accounting earnings & profitability variables have been at the top of these issues. Therefore earnings quality has been one of the important measures in the decision-making process of investors in field of bankruptcy prediction. This study has attempted to compare the prediction power of profitability variables among high quality and low quality earnings of Tehran Stock Exchange(TSE) companies and examine the effect of earnings quality on the efficiency of profitability variables in predicting the bankruptcy. In a sample of TSE companies, using artificial neural networks we find that the predictive accuracy of artificial neural networks for high quality earnings companies is significantly greater than of firms low quality earnings. Key Words: Profitability, Predictability, Earning Quality, Bankruptcy, Artificial Neural Network. Manuscript profile -
Open Access Article
296 - Designing non-linear pattern contagious influence of the Tehran Price Index from the physical assets market (Application of NARX artificial neural network model)
mahdi shaban habibollah nakhaei Ghodrat Alloh Talebnia nazanin bashirimaneshThe present study examines the contagiousness of the Tehran Stock Exchange from the price of parallel assets using the dynamic neural network. To perform calculations, the time series of coin price variables as a representative of the gold market, the average price per MoreThe present study examines the contagiousness of the Tehran Stock Exchange from the price of parallel assets using the dynamic neural network. To perform calculations, the time series of coin price variables as a representative of the gold market, the average price per square meter of residential building as a representative of the housing market. The price of each barrel of Iranian crude oil and the US dollar exchange rate and their conditional fluctuations as explanatory variables and the total index of Tehran Stock Exchange and its conditional fluctuation as the target variable from 1387 to 1397 are examined daily .The dynamic neural network is evaluated with four input variables and one target variable with different neurons with the MSE criteria, and the models with 20 neurons and 10 neurons have the lowest MSE, .Research results show that the stock exchange has a maximum of two lag from competing markets has become contagious, indicating the poor performance of the Tehran Stock Exchange. The results show that the proposed neural network patterns have a high power in predicting the index of Tehran Stock Exchange and its fluctuations from 1387 to 1397 as in-sample forecast and in 1398 as extra-sample forecast. Manuscript profile -
Open Access Article
297 - Designing an evaluation model for credit rating of Islamic securities with a Adaptive Neuro-Fuzzy network approach
Mohammad Shabani varnami Hosein Didehkhani Ali khozain arash naderianThe purpose of this research is designing a credit rating model for issuers and tools for financing Islamic securities in the Iranian capital market. To do this, three major steps were taken. The first step was to identify the evaluation criteria or the risks associated MoreThe purpose of this research is designing a credit rating model for issuers and tools for financing Islamic securities in the Iranian capital market. To do this, three major steps were taken. The first step was to identify the evaluation criteria or the risks associated with the Islamic securities, which was carried out by the experts and a review of theoretical basics. The second step, is modeling of Islamic securities using adaptive-network-based fuzzy approach in which the mean error of the training of all main and subset models was below the threshold. The third step is to apply adaptive fuzzy neural network modeling in credit rating of Islamic securities. In order to do this, the issuer’s ranking was used in the first stage and the results of the research showed that the issuer of the government had the least risk and private companies had the highest risk. In the second stage, for ranking financial instruments, the results showed that for issuer of government, treasury bonds had the lowest risk and forward bonds had the highest risk. For the issuer of state-owned companies, the forward bonds had the highest risk and lease bonds had the lowest risk. Manuscript profile -
Open Access Article
298 - Stock Price Prediction in Tehran Stock Exchange Using Artificial Neural Network Model and ARIMA Model: A Case Study of Two Active Pharmaceutical Companies in Stock Exchange
Ahmad Chegeni AZIZ GORDIn This Study We Compare the Efficiency of Both Artificial Neural Network Prediction Methods (ANN) and Traditional Method of Auto Regressive Integrated Moving Average (ARIMA) in Predicting Stock Prices in Iranian Stock Market. For This Purpose, Four Pharmaceutical Compa MoreIn This Study We Compare the Efficiency of Both Artificial Neural Network Prediction Methods (ANN) and Traditional Method of Auto Regressive Integrated Moving Average (ARIMA) in Predicting Stock Prices in Iranian Stock Market. For This Purpose, Four Pharmaceutical Companies, Alborz Drug, Iran Drug, Pars Drug, and Jam Drug Were Selected and ARIMA Model and Artificial Neural Network Model Were Estimated For All Four Companies. In Order to Estimate Artificial Neural Network Model, Stock Price Variable as Dependent Variable and Stock Trading Volume, Drug Industry Index, OPEC Oil Price, Exchange Rate and Gold Price are Considered as Independent Variables. MSE, RMSE, MAD, R2 and MAPE Criteria Were Used to Compare Two Models. In Order to Estimate the Stock Price Forecast Regression Model, Use of Auto Regressive Integrated Moving Average (ARIMA) Regression Is Used and Estimation of the Coefficients of the Model is Performed Using the EVIEWS Statistical Software. An Suitable ANN Model Was Created For Predicting Stock Prices Using MATLAB Software. The Results of the Research Showed That the Research Hypothesis is Correct and the Artificial Neural Network Model (ANN) Has a Better Predictor of Stock Price in the Iranian Stock Market Than the ARIMA Method. Manuscript profile -
Open Access Article
299 - Credit risk optimization model for crowdfunding process by using Neural Network(MLP)
ALI MALEKI Ali Zare Hashem NiKoumaram Shadi ShahverdianiThe purpose of this study is predict and design Credit risk model for debut crowdfunding .According, the complexity of the risk assessment the best neural network architecture with Customize hidden layer neurons selected Multilayer perceptron algorithm for simulation. T MoreThe purpose of this study is predict and design Credit risk model for debut crowdfunding .According, the complexity of the risk assessment the best neural network architecture with Customize hidden layer neurons selected Multilayer perceptron algorithm for simulation. The statistical population of this study is the financial information of credit / loan file of all customer (506 cases) one of the banks of the country for the year 1997-98. In order to show the significant relationship the extracted indices of the sample and the model output variables (non-default and default), the sample member tested by regression.Thus, thirteen indices entered to the model neural network input vector with three hidden layers in non-default and default groups. In the simulation results, the proposed model was able to optimize the weights of each of the inputs to the network with lower prediction error and 94.1% efficiency .also the average error absolute value obtained for training data (0.88), test data (0.94) and evaluation data (0.84) indicating high capability of the proposed model. According to the research Results, among the indices, income, 0.163 weight, Current Account weight 0.123 are more important, but “degree of education of education” 0.053 are less important in the non-defaulted group. Manuscript profile -
Open Access Article
300 - Financial Bankruptcy prediction using artificial neural network and firefly algorithms in companies listed in Tehran Stock Exchange
Mahdi Heidary Shokrollah Ziari seyed ahmad shayan nia Alireza Rashidi KemijanBy anticipating financial turmoil, it is possible to take the necessary precautions before financial distress occurs by managers and investors. This study compares two algorithms for prediction of bankruptcy using Artificial Neural Network (ANN) and Neural network optim MoreBy anticipating financial turmoil, it is possible to take the necessary precautions before financial distress occurs by managers and investors. This study compares two algorithms for prediction of bankruptcy using Artificial Neural Network (ANN) and Neural network optimized metaheuristic Firefly Algorithm (FA). To run test, first initial values are set for the network weights and biases and then during the optimization process, a population of different weights and biases is generated by FA algorithm. The conversion function used in the output layer is linear and for the middle layer a non-linear sigmoid function is selected. To conduct this research, the data of 79 companies listed on TSE during 2012 to 2015 were collected and analyzed statistically by backpropagation neural network and FA algorithms. The results show that FA, compared to ANN predicted the companies’ bankruptcy much better. Also, FA Algorithm maintains a good correlation between bankrupt and non-bankrupt companies, just like real data. Manuscript profile -
Open Access Article
301 - Modeling for Measuring Corporate Financial Sustainability Using the Econophysics and Bayesian Method
moloud soleimani Faegh ahmadi Mohammad Hossein Ranjbar Hamid Reza vakilifardThe concept of financial sustainability has been in the economic literature for nearly two decades. In the theoretical financial literature, firm financial sustainability can be described as a financial system consisting of financial intermediaries, markets, and market MoreThe concept of financial sustainability has been in the economic literature for nearly two decades. In the theoretical financial literature, firm financial sustainability can be described as a financial system consisting of financial intermediaries, markets, and market infrastructures that are capable of withstanding risk shocks and resolving financial imbalances. Therefore, according to the above argument, the purpose of the present study is to present a proposed model for measuring the financial sustainability of the company by using Econophysics and artificial neural network using a sample of 132 companies listed in Tehran Stock Exchange during 2015-2019. The results of the first hypothesis show that the prediction of financial sustainability based on the Econophysics method can provide better results. Also, based on the result of the second hypothesis, the Bayesian method can predict better financial sustainability. Finally, by comparing the Econophysics and Bayesian approaches to predicting financial sustainability of the firm, it can be concluded that the prediction of financial sustainability based on the economophysical method yields better results than the Bayesian method. Manuscript profile -
Open Access Article
302 - Stock price forecasting using a hybrid model based on recurring neural network and ANFIS and fuzzy expert system
Mostafa Yousofi Tezerjan Azam dokht Safi Samghabadi Azizollah MemarianiStock price forecasting is a challenging and attractive topic. Investors are interested in being able to predict the price of different stocks in financial markets. This paper presents a hybrid model that predicts the final stock price for the next day based on the adap MoreStock price forecasting is a challenging and attractive topic. Investors are interested in being able to predict the price of different stocks in financial markets. This paper presents a hybrid model that predicts the final stock price for the next day based on the adaptive neuro-fuzzy inference systems (ANFIS) and Return Neural Network (RNN) algorithm using historical data and indicators. Then the results of this model and the status of market rumors enter the fuzzy expert system based on the output of the fuzzy neural system and the return neural network along with the market rumor status and finalize the forecast. The combined model proposed to predict the stock price data of Mobarakeh Steel Company of Isfahan was implemented. In this study, for research data, the data of Tehran Stock Exchange Company related to the stock data of Mobarakeh Steel Company of Isfahan from April 26, 2016 to March 20, 2017 has been used. Four technical indicators used in this study are: Moving Average(MA), Exponential Moving Average(EMA), Relative Strength Index(RSI), and Moving Average Convergence Divergence(MACD). These variables have been used as the input of the adaptive neuro-fuzzy inference systems(ANFIS) to predict the final price of the next day's shares. Manuscript profile -
Open Access Article
303 - Presentation of intelligent Meta-heuristic Hybrid models (ANFIS -MGGP ) to predict stock returns with more accuracy and speed than other Meta-heuristic methods.
mahmood kohansal kafshgari Alireza Zarei reza behmaneshDiscussions about forecasting Stock returns in developed countries has long been regarded as one of the most interesting scientific topics.However,due to many problems,the correct prediction of stock returns has remained a matter of strengthTtherefore,the researcher see MoreDiscussions about forecasting Stock returns in developed countries has long been regarded as one of the most interesting scientific topics.However,due to many problems,the correct prediction of stock returns has remained a matter of strengthTtherefore,the researcher seeks to provide an accurate,practical and effective model for predicting stock returns for investors.The statistics sampel of research is consist of 138 active companies in Tehran Stock Exchange from 2008 to 2017 wich are selected by the systematic removal method . ANFIS,MGGP, regresion and neural network and different statistics tests are used for data analysis. For impelement of these techniques MATLAB and GenXproTools software are used respectively.The result of the study showed that in oreder to predict stock returns.the use of a meta –heuristic Hybrid models is more accurate and faster than other meta huristic models.Because ,first the most optimal input variables are selected through the ANFIS technique and then predicted using theMGG meta heuristic model.Therefore,due to the correct choice of input variables,predicting stock returns is both more accurate and faster.In addition ,the mathematical model is used to predict. Manuscript profile -
Open Access Article
304 - Designing a Model for Forecasting the Gold Price Returns (Emphasizing on Combined convolutional neural network Models and GARCH Family Models)
Mohammad Javad Bakhtiaran mehdi ZolfaghariFinding the best way to optimize the portfolio is one of the concerns of activists in the investment management industry. In recent years, the introduction of economic and mathematical models in the prediction of Gold indice has helped many investors to optimize portfol MoreFinding the best way to optimize the portfolio is one of the concerns of activists in the investment management industry. In recent years, the introduction of economic and mathematical models in the prediction of Gold indice has helped many investors to optimize portfolios. Therefore, in this study, we introduce models of GARCH family composition and convoultional neural network to predict the daily yield of Gold index will be paid during the period of 1390-1398. In this study, the Gold index is examined using GARCH and EGARCH short-term memory models. Of the two variables, the price of crude oil and the dollar index as factors that their shocks and fluctuations have a major impact on Gold indices are used as control variables. In addition to using convolutional model, considering the better performance of combined models (compared to individual models ) In anticipation In this study, all models of the GARCH family (both short and long run) with the convoultional neural network were combined and using the combined models, the efficiency of the main stock index and the five selected indicators for the next 10 days were predicted step by step and its accuracy Based on the evaluation criteria. Manuscript profile -
Open Access Article
305 - Financial Innovation Test in Banking: Providing a Hybrid Model for Forecasting and Assessing Credit Risk of Medium and Small Enterprises (SMEs) in Commercial Banks
Kokab Sharifi Amir Mohammadzadeh Hashem Nikoumaram Naser HamidiWe live in an age characterized by the very rapid rate of financial innovation. The study of the historical evolution of progress and economic development of developed and industrialized countries shows that one of the main factors in the emergence of rapid and massive MoreWe live in an age characterized by the very rapid rate of financial innovation. The study of the historical evolution of progress and economic development of developed and industrialized countries shows that one of the main factors in the emergence of rapid and massive growth has been the existence of financial reforms in these countries. There are different incentives for individuals and active enterprises in the financial system to perform financial innovation, which is one of the most important incentives, the introduction of tools and methods to reduce, eliminate or manage existing risks. One of the most important tools the current situation that can help banks in the optimal management of consumption and prevention of claims. Designing and applying credit risk assessment models in granting facilities. The purpose of this study is to provide a suitable model for financial innovation based on credit risk measurement of SMEs in commercial banks. In this regard, effective indicators on the credit risk of SMEs were identified by using the genetic algorithm method and logit, neural network and fuzzy expert system were evaluated. The results show that the using the hybrid model has more accurate results in the assessment the credit risk of SMEs. Manuscript profile -
Open Access Article
306 - Comparison of different machine learning models in stock market index forecasting
maryam sohrabi Seyed Mozaffar mirbargkar Ebrahim Chirani SINA KHERADYARPredicting time series of financial markets is a challenging issue in the field of specialized studies of time series and has attracted the attention of many researchers. Due to the presence of big data, this issue has led to the growth of developments in the field of m MorePredicting time series of financial markets is a challenging issue in the field of specialized studies of time series and has attracted the attention of many researchers. Due to the presence of big data, this issue has led to the growth of developments in the field of machine learning models. Due to the importance of this issue, in this study, by using the comparison of different machine learning models such as random forest approaches, support vector machine, artificial neural network and deep learning-based recurrent neural networks to investigate the ability of different machine learning models in prediction. The total index of Tehran Stock Exchange during the period 2013 to 2020 has been discussed. The prediction results of 1, 3 and 6 day courses for the out-of-sample period show that the machine learning method based on the long short-term memory (LSTM) network, a recurrent neural networks, has a better result compared to other models. Manuscript profile -
Open Access Article
307 - The Assessment of the optimal Deep Learning Algorithm on Stock Price Prediction (Long Short-Term Memory Approach)
Amir Sharif far Maryam Khalili Araghi Iman Raeesi Vanani Mirfeiz FallahForecasting stock prices plays an important role in setting a trading strategy or determining the appropriate timing for buying or selling a stock. Deep Learning (DL) is a type of Artificial Neural Network (ANN) that consists of multiple processing layers and enables hi MoreForecasting stock prices plays an important role in setting a trading strategy or determining the appropriate timing for buying or selling a stock. Deep Learning (DL) is a type of Artificial Neural Network (ANN) that consists of multiple processing layers and enables high-level abstraction to model data. The key advantage of DL models is extracting the good features of input data automatically using a general-purpose learning procedure which is suitable for dynamic time series such as stock price.In this research the ability of Long Short-Term Memory (LSTM) to predict the stock price is studied; moreover, the factors that have significant effects on the stock price is classified and legal and natural person trading is introduced as an important factor which has influence on the stock price. Price data, technical indexes and legal and natural person trading is used as an input data for running the model. The results obtained from LSTM with Dropout layer are better and more stable than simple form of LSTM and RNN models. Manuscript profile -
Open Access Article
308 - Predicting the daily index of the Tehran Stock Exchange using the selection of appropriate features for the Long Short-Term Memory neural network (LSTM)
Somayeh Mohebi Mohammad Esmaeil Fadaeinejad mohammad osoolian Mohammad reza HamidizadehThe stock market index is one of the effective features in investment because it can well reflect the health status and macro change trend of a country’s economic development. Various features affect the stock index. The various combinations of these features crea MoreThe stock market index is one of the effective features in investment because it can well reflect the health status and macro change trend of a country’s economic development. Various features affect the stock index. The various combinations of these features create a wide state space. Hence, it is impractical to provide a data set containing all these combinations to train the stock index prediction model. in this research, an attempt has been made, after collecting a significant number of effective features on the index, to provide a method for selecting appropriate features for the stock index prediction model with aim of increasing prediction accuracy. For this purpose, the mRMR algorithm is used as the basic algorithm. Also, to select the appropriate model, a number of the most applicable artificial intelligence models for predicting the stock index were compared and according to the results, the LSTM network was selected to predict the stock index. The results of this study show that using the LSTM network and the proposed method in selecting features, with 8 selected features, high accuracy can be achieved in the daily prediction of the Tehran Stock Exchange Index. So that MPE is calculated to be about 2.66, Manuscript profile -
Open Access Article
309 - Comparison of the Predictive Accuracy of Artificial Neural Network Systems Based on Multilayer Perceptron Approach and Falmer Binary-Logistics Model in Order to Predict Bankruptcy
Somieh Saroei Hamid Reza Vkili Fard, Ghodratolah Taleb NiaFinancial analysts and other users need relevant and reliable information to predict corporate bankruptcy, which should be distributed symmetrically to all users. Accordingly, the purpose of this study is to compare the prediction accuracy of Artificial Neural Network ( MoreFinancial analysts and other users need relevant and reliable information to predict corporate bankruptcy, which should be distributed symmetrically to all users. Accordingly, the purpose of this study is to compare the prediction accuracy of Artificial Neural Network (ANN) systems based on the Multilayer Perceptron Approach and Falmer Binary-Logistics Model in order to predict bankruptcy. To test the hypotheses, the combined data of 172 companies listed on the Tehran Stock Exchange in the period 2007-2016 were used. The results of the analysis of the research data show that the ANN system can identify of the factors affecting on bankruptcy of Iranian companies in the year before bankruptcy by Precision equal 98%. Findings from the binary-logistic model showed that the forecasting model designed based on the Falmer regression method is able to predict with 82% accuracy the bankruptcy of the sample companies. Therefore, the use of artificial neural networks can more powerfully and accurately predict bankruptcy than regression models. Manuscript profile -
Open Access Article
310 - Provide IPO valuation model using genetic algorithm and compare the value of the proposed model with Op
samaneh fathalian seyed Ali Nabavi Chashmi Ebrahim ChiraniProper Ipo Valuation Companies entering the capital market for the first time are critical to both business owners and investors. But the valuation of these stocks is influenced by many quantitative and qualitative factors. Nonlinear intelligent systems such as neural n MoreProper Ipo Valuation Companies entering the capital market for the first time are critical to both business owners and investors. But the valuation of these stocks is influenced by many quantitative and qualitative factors. Nonlinear intelligent systems such as neural networks and genetic algorithms are good tools for accurately predicting the initial stock value. Therefore, the purpose of this study is to present the IPO valuation model using genetic algorithm and compare the value of the proposed model with Op. For this purpose, data related to 421 companies were collected that during the years 2009 to 1397 had made an initial public offering of shares on the Tehran Stock Exchange. In order to analyze the data, the methods of regression, neural network and genetic algorithm have been used. The results showed that the Ipo valuation model using genetic algorithm is the optimal IPO valuation model. Also, the projected valuation, while close to the OP, while meeting the relative price increase, can meet the expectations of investors and business owners in a proper IPO valuation. Manuscript profile -
Open Access Article
311 - Explain the factors affecting stock liquidity using genetic algorithm and minimum and maximum correlation (MRMR) methods
Mahmoud Rezaei Hossein Panahian Mahdi Madanchi Zaj Hasan GhodratiLiquidity of stocks is an important challenge in the capital market. Identifying the factors affecting liquidity helps to predict the stock liquidity situation and thus stock risk management. The purpose of this study is to find the factors affecting the liquidity of st MoreLiquidity of stocks is an important challenge in the capital market. Identifying the factors affecting liquidity helps to predict the stock liquidity situation and thus stock risk management. The purpose of this study is to find the factors affecting the liquidity of stocks. For this purpose, in the first stage, using the research literature and experts, the influencing factors are identified and using the methods of minimum redundancy and maximum correlation (MRMR) and genetic algorithm, the effective variables are selected. In this research, using Excel software and existing raw data, the required data was created and then using support software and neural network toolbox and support vector machine was created. . Finally, the extracted variables using MRMR include stock market value, intensity of product market competition, GDP growth, equity returns, stock returns, inflation rate and family ownership, and using the financial model of financial leverage, government ownership, Equity returns, GDP growth, share buoyancy percentage, market type and board (on the stock exchange and OTC), the intensity of competition in the product market were selected. Manuscript profile -
Open Access Article
312 - Price predicting with LSTM artificial neural network and portfolio selection model of financial assets and digital currencies
Faranak Khonsarian Babak teimourpour Mohammad Ali RastegarFinding solutions for price prediction, forming an optimal portfolio and achieving more profit are the basic goals of financial market activists. The purpose of this research is to predict the price of financial assets such as several stocks, gold, coin and a number of MoreFinding solutions for price prediction, forming an optimal portfolio and achieving more profit are the basic goals of financial market activists. The purpose of this research is to predict the price of financial assets such as several stocks, gold, coin and a number of digital currencies using the LSTM neural network model and then form an optimal portfolio by calculating the rate of return, risk and the Sharpe ratio. The data used is from the archives of the Tehran Stock Exchange website, the website of the gold, coin and currency information network, as well as the website of buying and selling digital currencies. The time series of the prices of the investigated assets is between 2017 and 2020. Also, we used Python programming language and Gephi software to build the model and analyze the data. In the end, it was found that the LSTM neural network model is capable of predicting the price of financial assets with a very low error rate in each asset, and according to the Sharpe ratio obtained for each financial asset and the correlation matrix, Vebank stock, Khbahman 1 stock, and Digital currencies TRON, Tether and Bitcoin allocate more shares in the proposed portfolio. Manuscript profile -
Open Access Article
313 - Development of stock portfolio trading systems using machine learning methods
Ali Heidarian Mohadeseh Moradi Mehr Ali FarhadianInvestment portfolio theory is an important foundation for portfolio management, which is a well-studied but not saturated topic in the academic community. Integrating return forecasting in investment portfolio formation can improve the performance of portfolio optimiza MoreInvestment portfolio theory is an important foundation for portfolio management, which is a well-studied but not saturated topic in the academic community. Integrating return forecasting in investment portfolio formation can improve the performance of portfolio optimization model. Since machine learning models have shown a superiority over statistical models, in this research, a approach of forming the stock portfolio in two stages is presented. first step, by implementing neural network, suitable stocks are selected for purchase, in the second step, using the (MV) model, the optimal weight in investment portfolio is determined for them. In particular, the stages of selecting suitable stocks and forming a stock portfolio are the two main stages of the model developed in this research. first step, a convolutional neural network model is proposed to predict stock buy and sell points for the next period.second step, stocks that are labeled as buys are selected as stocks suitable for buying, and MV model is used to determine their optimal weight in the stock portfolio. The results obtained using 5 shares of Tehran stock market as a study sample show that the efficiency and Sharpe ratio of proposed method is significantly better than traditional methods (without filtering suitable stocks) Manuscript profile -
Open Access Article
314 - Forecasting the bank's financial resources using the linear model (ARIMA) and nonlinear artificial fuzzy networks
omid mehrinamakawarani reza ehteshamrasiOne of the most important issues of banking managers as an influential variable on the banking industry is the knowledge of the status of bank deposits that the bank depends on a large extent on it. Therefore, bank managers are keen to know how much the total bank depos MoreOne of the most important issues of banking managers as an influential variable on the banking industry is the knowledge of the status of bank deposits that the bank depends on a large extent on it. Therefore, bank managers are keen to know how much the total bank deposits will be at a given time in the future. Predicting the amount of deposits, changes and fluctuations of these deposits can help banks in planning and decision making. In this research, using statistical techniques and approach of artificial neural network models, we have tried to introduce a model with the highest estimation power and the least amount of error to predict the amount of deposits or the same sources of finance by their different types for the desired bank. To test the hypotheses, one private bank information was used during the period of 1387-1396. In this research, we compared the predictive power of ARIMA and artificial neural network method. To assess the accuracy of forecasting the bank's resources, the ARIMA method used Coopiff and Christopherson tests. The results of the research on the amount of bank deposits monthly showed that the neural network method provides better estimates than the ARIMA method. Manuscript profile -
Open Access Article
315 - مقایسه مدل های شبکه عصبی با مدل سری زمانی باکس- جنکینز در پیش بینی شاخص کل قیمت سهام بورس اوراق بهادار تهران
جلال حقیقت منفرد محمود احمدعلینژاد سارا متقالچی -
Open Access Article
316 - ارائه مدلی برای شناسایی عوامل موثر بر قیمت آتی سکه به روش شبکه عصبی مصنوعی و مقایسه آن با مدلهای رگرسیونی
میلاد گودرزی بهزاد امیری -
Open Access Article
317 - بررسی توان تبیین مدل های شبکه عصبی درسنجش میزان ارزش در معرض خطر
فرهاد غفاری هاشم نیکومرام غلامرضا زمردیان -
Open Access Article
318 - رویکرد شبکه عصبی مبتنی بر کلونی زنبور عسل مصنوعی
سعید فلاح پور رضا راعی محمد هندیجانی -
Open Access Article
319 - مقایسه عملکرد مدلهای رگرسیونی ARIMA وشبکه عصبی باالگوریتم ژنتیک (GMDH) درپیش بینی قیمت نفت خام ایران
عباسعلی ابونوری ناهید خدادادی -
Open Access Article
320 - طراحی مدل پیش بینی ورشکستگی شرکت ها به وسیله شبکه های عصبی فازی (مطالعه موردی:شرکت های بورس اوراق بهادار تهران)
مریم ظهری محمدعلی افشارکاظمی -
Open Access Article
321 - مقایسه توان تبیین مدلهای پارامتریک (اقتصادسنجی) و شبکه عصبی در سنجش میزان ارزش درمعرض خطر پرتفوی شرکت های سرمایه گذاری جهت تعیین پرتفوی بهینه در بازار سرمایه ایران
غلامرضا زمردیان علی رستمی مهدی کریمی زند -
Open Access Article
322 - Providing a model for predicting stock prices using ultra-innovative neural networks
Seyyed Hosein Miralavi zahra pourzamaniDue to the complexity of the stock market and the high volume of processable information, often using a simple system to predict cannot release appropriate results. Therefore, researchers have been trying to provide a system with less complexity and more efficiency and MoreDue to the complexity of the stock market and the high volume of processable information, often using a simple system to predict cannot release appropriate results. Therefore, researchers have been trying to provide a system with less complexity and more efficiency and accuracy using hybrid models. nowadays various patters are used including statistical technique (discriminate analysis , logistic , analysis factors) and artificial intelligent techniques ( neural networks(NN) , decision trees , case based reasoning , genetic algorithm , rough sets , support vector machine , fuzzy logic ) and the combination of these two technique for predicating stock prices. For most predictive models, the system uses only one indicator to predict, but in the proposed model in this study, a two-level system of multilayered perceptron neural networks is presented which uses several indicators to predict. To do this, required information of Tehran Stock Exchange price indicators, for fiscal years 2012 - 2017 was collected. We also used the Grasshopper Optimization Algorithm to select the best samples for better nerve network training and thus to improve the results. The results show that the proposed model can operate with lower prediction error than other models. Manuscript profile -
Open Access Article
323 - مقایسه توان تبیین مدل های ناپارآمتریک و مدل های شبکه عصبی در سنجش میزان ارزش درمعرض خطر پرتفوی شرکت های سرمایه گذاری جهت تعیین پرتفوی بهینه در بازار سرمایه ایران
غلامرضا زمردیان -
Open Access Article
324 - پیش بینی بازده شاخص بورس اوراق بهادار با استفاده از مدلهای شبکه ها عصبی مصنوعی شعاع پایه
رضا تهرانی سعید مرادپور -
Open Access Article
325 - کاربردالگوریتم ژنتیک خطی و غیر خطی در بهبود قدرت پیشبینی
زهرا پورزمانی -
Open Access Article
326 - مقایسه عملکرد مدل فاما و فرنچ و شبکه های عصبی مصنوعی
ناصر شمس سمیرا پارسائیان -
Open Access Article
327 - Forecast earnings management based on adjusted Jones model using Artificial Neural Networks and Genetic Algorithms
Khosro Faghani Makrani S. Hasan Salehnezhad Vahid AminIn recent years, earnings management in university research has attracted much attention. The aim of this study is to predict earnings management through discretionary accruals based on adjusted Jones model. In this study, two models of artificial neural networks and ge MoreIn recent years, earnings management in university research has attracted much attention. The aim of this study is to predict earnings management through discretionary accruals based on adjusted Jones model. In this study, two models of artificial neural networks and genetic algorithms - neural network hybrid model as a successful model to predict earnings management based on adjusted Jones model were used in the Tehran Stock Exchange. The sample used in this study is consisted of 570 firm-year between 2008 to 2013. The results showed that neural networks have a high ability to predict earnings management rather than the adjusted Jones linear model. The findings also suggest that, the genetic algorithm through optimizing artificial neural network weights is able to increase power of artificial neural network to predict earnings management. Manuscript profile -
Open Access Article
328 - پیشبینی بازده صندوقهای سرمایهگذاری مشترک
کاظم چاوشی ابراهیم صابر -
Open Access Article
329 - Optimization of technical indicators’ parameters for intraday data using optics – inspired optimization (OIO): a case study of Tehran stock exchange
Mohammad Ali Rastegar Farah AshuriIn this paper a stock trading system based on the combination of six technical indicators is designed. The indicators are combined using an artificial neural network and their parameters are optimized using convex combination-based optics-inspired optimization (COIO) al MoreIn this paper a stock trading system based on the combination of six technical indicators is designed. The indicators are combined using an artificial neural network and their parameters are optimized using convex combination-based optics-inspired optimization (COIO) algorithm. In the proposed model the technical indicators’ optimized parameters are obtained using both COIO and genetic algorithms with the aim of maximization of modified Sharpe ratio. The presented paper uses stock intra-day prices as input data and considers the transaction costs. The designed strategy is compared against several other approaches including: using the indicators’ default parameters, buy and hold strategy and optimization using genetic algorithm, for both daily and intra-day prices and due to a greater modified Sharpe ratio for the proposed model, its superiority is shown in all cases. Moreover, in a comparison based on end- of- period returns, it is shown that without considering the transaction costs the results of the intra-day data beats the results of the daily data while no superiority is observed when considering the transaction costs. So reducing the transaction costs is recommended to motivate traders to trade on an intra-day basis. Manuscript profile -
Open Access Article
330 - Forecasting Stock Price Trend by Artificial Neural Networks (Case Study: Isfahan Oil Refinery Company)
hossein badiei Ruhollah Rezazadeh Hadi MahmoudiArtificial neural networks (ANN) are mathematical models inspired by human’s neural and brain system. This research deals with the next day price forecasting in Tehran’s stock market by MLP, and attempts, by various methods, to reduce the prediction error. H MoreArtificial neural networks (ANN) are mathematical models inspired by human’s neural and brain system. This research deals with the next day price forecasting in Tehran’s stock market by MLP, and attempts, by various methods, to reduce the prediction error. High pricing of stocks may lead to low demand for negotiable stocks and the failure of privatization. Raising various doubts in the negotiation of public properties, low pricing results in the long-term failure of negotiation policies. With respect to the importance of this issue, the newness of stock market and the lack of financing institutes and investment banks in Iran, prediction of stock price trend and its ascending and descending order can influence the decisions and strategies of managers. Various variables affect stock prices among which the role of economic indices, such as exchange rate / oil price and gold price is significant. The purpose of the present study is to predict the final prices of stocks by utilizing daily data through neural networks. The results indicate that the ANN model has low error and high explanatory and thus considerable forecasting power. Manuscript profile -
Open Access Article
331 - Neuro-Genetic Structure to valuation of Initial Public Offering
ali rostami Emad Falamarzi sara FaroughiConsidering stock market history, major concerns in the first phase to enter the capital market is that what the right price for the initial public offering and could they convince investors to buy shares. Besides that, there are also investors concerns about the accura MoreConsidering stock market history, major concerns in the first phase to enter the capital market is that what the right price for the initial public offering and could they convince investors to buy shares. Besides that, there are also investors concerns about the accuracy of the pricing stocks. This study uses nonlinear method has resolved this issue. Study provides a model pricing initial public offering of shares on the Tehran Stock Exchange. The research period between 1382 to 1393. Research population 145 enterprises entering the Tehran Stock Exchange in this period of time and the sample of study is according to the condition of the Company and continuous investment of funds and access to company data, were reduced to 103 companies. The proposed network is a neural network optimized the genetic algorithm to determine the price of shares of new companies entering the stock exchange.With a choice of 12 variables affecting the price of initial public offerings and one dependent variable (Initial Public Offering price) suitable model to _ pricing than other linear models presented. The results of the fourth measure, RMSE, MAE, R-SQUARE, U-THEIL reflect the correct pricing proposed model, in most cases. Manuscript profile -
Open Access Article
332 - کاربرد شبکه عصبی- فازی انطباقی در پیشبینی قیمت سهام شرکت ایرانخودرو
ابراهیم عباسی امیر ابوئی مهریزی -
Open Access Article
333 - Stock price prediction based on LM-BP neural network and over-point estimation by counting time intervals: Evidence from the Stock Exchange
Mohammadreza Vatanparast masoud asadi Shaban Mohammadi abbas babaeiIn this study, to determine the stock price forecasting method, a LM-BP neural network was presented based on time series with respect to open price, highest price, lowest price, package price and volume of transactions. In the present study 315 days of stock prices wer MoreIn this study, to determine the stock price forecasting method, a LM-BP neural network was presented based on time series with respect to open price, highest price, lowest price, package price and volume of transactions. In the present study 315 days of stock prices were chosen to create 10 samples and the test set includes stock prices from day 316 to day 320 and used the LM-BP neural network. In this research, the determination of the critical point of excess, asymmetry and counting of intervals were investigated. The curve MRE2-MRE1 was plotted and the precision related to the best prediction of the BP neural network was estimated based on several independent replicas. The post-test was performed using a Kupiec Test and a Christopherson test. The results showed that stock price prediction based on the LM-BP neural network and over-point estimation by counting the intervals resulted in better results than the existing methods. Manuscript profile -
Open Access Article
334 - Explaining the Optimal Model of Appraisal and Pricing of the Initial Pablic Offering using Fuzzy Multi-Criteria Decision Making Techniques, Multivariate Regression, Neural Network and Genetic Algorithm
samaneh fathalian sayyed Ali Nabavi Chashmi Ebrahim ChiraniValuation and pricing of securities the process of estimating the value of securities is one of the initial shares of companies. Because, on the one hand, investors need to know in a conscious investment that knows the true value of the stock they are interested in inve MoreValuation and pricing of securities the process of estimating the value of securities is one of the initial shares of companies. Because, on the one hand, investors need to know in a conscious investment that knows the true value of the stock they are interested in investing in and on the other hand, the owners of companies that are going to sell their securities have to evaluate and value their assets in a proper manner. Therefore, the purpose of this study is to explain the optimal model of evaluation and pricing of the initial public supply of shares of companies accepted in Tehran Stock Exchange using fuzzy multi-criteria decision-making techniques, stepwise regression, neural network and genetic algorithm. To this end, data on 421 companies were collected that during the years 2006 to 2018 launched a public offering of shares on the Tehran Stock Exchange. Fuzzy AHP method, forward regression, neural network and genetic algorithm are also used to analyze the data. The results of the research showed that the genetic algorithm model is the optimal pricing model and initial stock valuation. Manuscript profile -
Open Access Article
335 - Predict the risk of falling stock prices by using meta-innovative methods (Cumulative particle motion optimization algorithm) and comparison with logistic regression
Esfandiar Malekian hossin fakhari jamal ghasemi Sarveh FarzadThe Crash, which indicates how much specific stock prices are at risk of collapse. Accordingly, the purpose of this research is to model the risk of falling stock price of listed companies in Tehran Stock Exchange using a multivariate optimization algorithm for particle MoreThe Crash, which indicates how much specific stock prices are at risk of collapse. Accordingly, the purpose of this research is to model the risk of falling stock price of listed companies in Tehran Stock Exchange using a multivariate optimization algorithm for particle cumulative movement and comparing results with logistic regression. For this purpose, a hypothesis was developed for the study of this issue and the data for 106 members of the Tehran Stock Exchange for the period of 2010-2010 were analyzed. First, 14 independent variables were introduced as inputs of the combined genetic algorithm and artificial neural network, which was considered as a feature selection method, and 7 optimal variables were selected. Then, using particle cumulative algorithm and logistic regression, predicted The risk of falling stock prices. The stock price collapse criterion has been used to calculate the risk of falling stock prices. The research findings show that the particle agglomeration algorithm is more likely than traditional logistic regression to predict the risk of falling stock prices. These findings underscore the need for managers to use meta-metric methods for forecasting. Manuscript profile -
Open Access Article
336 - A Comparison between Fama and French five-factor model and artificial neural networks in predicting the stock price
reza tehrani Milad Heyrani Samira MansuriOne of the most important issues of financial markets is the prediction of price and stock returns. In this paper, we try to find the best model and stock price prediction approach based on the mean square error (MSE), root-mean-square error (RMSE), R-squared, standard MoreOne of the most important issues of financial markets is the prediction of price and stock returns. In this paper, we try to find the best model and stock price prediction approach based on the mean square error (MSE), root-mean-square error (RMSE), R-squared, standard deviation (SD), Mean absolute error and the mean absolute percent error (MAPE) for the Fama and French five-factor model. For this purpose, after the formation of a portfolio based on the Fama and French model during the period from 2009 to 2017, stock price is estimated by econometric model, neural network and Fuzzy Neural Networks, so the accuracy of each approach was compared. The results of the prediction the efficiency of the generated portfolios show that the prediction accuracy of the radial base function network (RBF) is very high compared to other ARMA models and other neural networks. Manuscript profile -
Open Access Article
337 - Application of methods radial neural network, Gaussian process regression in predicting financial constraints Companies admitted to the Tehran Stock Exchange
Mohammadreza Gholamzadeh Mahdi faghani ahmad pifeOne of the important issues in predicting financial constraints is the choice of predictor variables. In this study, we investigated the Gaussian process machine learning method and radial neural network to predict financial constraints. For this purpose, 208 companies MoreOne of the important issues in predicting financial constraints is the choice of predictor variables. In this study, we investigated the Gaussian process machine learning method and radial neural network to predict financial constraints. For this purpose, 208 companies from 1390 to 1396 have been selected as the statistical population. Due to the availability of information, all companies have been considered as a statistical sample. The results of this study showed that machine learning methods have the ability to predict the financial constraints of corporations admitted to Tehran Stock Exchange. Therefore, the main hypothesis of this research is confirmed and machine learning methods are an effective way to predict financial constraints. Also, the results showed that the company's value, operating cash flow ratio, financial leverage, return on assets, and the percentage of institutional owners had the most importance in predicting financial constraints. Manuscript profile -
Open Access Article
338 - To Forecat the Recession and Prosperity in the Tehran Stock Exchange using Models of MS and NSGA-ANN
farzaneh abdollahian Mohammad Ebrahim Mohammad Pourzarandi Mohammad Hasheminejad Mehrzad MinoueiThe stock exchange is one of the financial instruments of countries around the world. The recession in this market can have important effects, for example reducing liquidity, reducing the profitability of companies admitted to the stock exchange, and reducing economic g MoreThe stock exchange is one of the financial instruments of countries around the world. The recession in this market can have important effects, for example reducing liquidity, reducing the profitability of companies admitted to the stock exchange, and reducing economic growth. In this paper, we are looking for extraction and prediction of time cycles in the stock market. Initially, using the total stock index and the MSI (3) AR (2) model, three cycles of recession, medium prosperity and high prosperity are extracted in the stock market. Then the most important predictor variables are determined by using the integration of the NSGA (II) algorithm and the three types of neural network models and predicted the market situation for the next three months. Finally, the performance of three types of multilayer perceptron neural network, radial basis and probable network were compared in terms of feature selection and prediction of future market situation. The results indicate that all three models have acceptable error rates, accuracy, and Kappa coefficients, and the probable network model has lower error rate, more accuracy and kappa coefficient than other models. Manuscript profile -
Open Access Article
339 - Interval Forcasting for Gold Price with hybrib model of ARIMA and Artificial Neural Network
Shapor Mohammadi Reza Raeie Mohammadreza RahimiPrice forecasting is one of the most challenging issues that the speculators, traders and brokers are faced with. On the other hand in interval analysis it is supposed that observations and estimations in the real world are not complete and reliable so to increase the a MorePrice forecasting is one of the most challenging issues that the speculators, traders and brokers are faced with. On the other hand in interval analysis it is supposed that observations and estimations in the real world are not complete and reliable so to increase the accuracy we should describe the data as the intervals that includes real quantities. Various methods are used in order to model the time series such as price. Autoregressive integration moving average (ARIMA), which is known as box-Jenkins method is one of the most commonly used models in forecasting of time series during the past three decades. But the main assumption is that there is a linear relationship between the values of the series therefore nonlinear relationships cannot be explained completely by using autoregressive integration moving average (ARIMA). Another method in time series forecasting is neural network which can estimate the various nonlinear relationship (called neural network universal estimating) but according to the literature, using network will have complicated results. Since it is difficult to understand the linear and nonlinear data pattern in reality, this idea will come to mind that the combination of linear and nonlinear models could increase the accuracy of forecasting. So in this research the linear part will be estimated by ARIMA and then the non-linear residuals will be modeled by neural network and finally the predicted result will be added to ARIMA in order to forecast the low, high and close price of gold .comparing the accuracy of the hybrid model to ARIMA and neural network by pair compared, Diebold-Mariano and Harvey-Newbold –Leybourn test and two criteria (MSE and MAE) showed that the hybrid model presented better performance. Manuscript profile -
Open Access Article
340 - Investigate the Operation of Random forest and Deep neural networks on Statistical Arbitrage Strategy
alireza Fazlzadeh Jafar Haghigha Faranak Pourkeivan vahid ahmadian In this research, the statistical analysis of random forest effects has been done. Also, to evaluate the performance of the random forest algorithm in the field of statistical arbitrage compared to other models presented in the previous research, the comparison of More In this research, the statistical analysis of random forest effects has been done. Also, to evaluate the performance of the random forest algorithm in the field of statistical arbitrage compared to other models presented in the previous research, the comparison of the results from the application of this algorithm with deep neural network algorithm has been done. The models are taught with stock price information and the output from this technique categorizes stocks according to the position of buying and selling. Using this strategy, profitable positions are identified in market shares for profit. The results showed that the model of random forest with less error classification than deep neural network model. Using this strategy, profitable positions are identified in market shares for profit. The results showed that the model of random forest with less error classification than deep neural network model. Manuscript profile -
Open Access Article
341 - Forecasting the exchange rate of euro to dollar with the artificial neural network technique
shafagh sharif moghadam seyyed Zabihollah HashemiExchange rate prediction is an important economic variable of interest to the economic actors. Technical approach is one of the commonly used approaches to forecasting, which uses the past behavior of the exchange rate for prediction. However, given the chaotic and non- MoreExchange rate prediction is an important economic variable of interest to the economic actors. Technical approach is one of the commonly used approaches to forecasting, which uses the past behavior of the exchange rate for prediction. However, given the chaotic and non-linear structure of financial markets, the market forecasting cannot be done using a certain and simple method obtained by combining different technical tools and more sophisticated methods are required. In recent decades, neural networks have been employed as one of the most widely used methods in classification, pattern recognition and prediction of complex time series. In this research, a multilevel neural network model was provided to predict the euro-dollar exchange rate, which predicts the price on the next day with an appropriate accuracy by utilizing the data and variables derived from the technical analysis. The results demonstrated the proper function of this method versus other conventional methods of technical analysis and neural network. Manuscript profile -
Open Access Article
342 - Presenting the developed model of Benish model with emphasis on the special characteristics of the company using neural network, vector machine and random forest
Kiumars Pourgadimi Saeed jabbarzadeh Kangarloui Jamal Bahri SalesAbstractAs the business process becomes more complex, the risk of financial statements being distorted increases with each passing day. In this regard, researchers have been looking for models to detect fraud in financial statements. Benish (1997) used a combination of MoreAbstractAs the business process becomes more complex, the risk of financial statements being distorted increases with each passing day. In this regard, researchers have been looking for models to detect fraud in financial statements. Benish (1997) used a combination of financial ratios and accruals to predict profit manipulation methods. since auditors are presented as external oversight in the corporate governance structure of the company's performance, in this study the model is developed based on the qualitative characteristics of the auditor, which include the auditor's size, auditor tenure, reporting delay, Auditor Class and Auditor Change.The fitting of the stochastic vector machine, random forest and neural network has been used to fit the extended model. The results show that the coefficients obtained from the random forest model are 99% and more than the two neural network and vector model 94%. Manuscript profile -
Open Access Article
343 - Predicting the success of the investment projects of Aras and Maku commercial-industrial free zones and Salmas special economic zone using perceptron multilayer neural network technique
morteza shokrzadeh mojtaba shokrzadehTo analyze the data of this research descriptive statistics and inferential statistics were used and experts selection software, MATLAB SPSS and PLS software were employed.Using theoretical foundations and libraries, six effective factors and variables predicting the su MoreTo analyze the data of this research descriptive statistics and inferential statistics were used and experts selection software, MATLAB SPSS and PLS software were employed.Using theoretical foundations and libraries, six effective factors and variables predicting the success or failure of Investment projects in the free and special economic zones of the country were identified. After describing the variables and testing the normality,using the PLS software, a confirmatory factor analysis of the variables was carried out, in which all of the factors had a good confirmatory factor analysis and all the questions were approvedThen, using linear regression and ANOVA, the effect of each of the factors on the success or failure of investment projects was investigated, and the results of this test showed confirmation of the impact of each of the factors, and then the results of the hierarchical analysis indicated this was the first rank of product and service, followed by the second-rank ,that is geographical considerations, and the characteristics of the investor's psychology, the third rank, the product market characteristics, the fourth rank, the investor's ability to rank fifth, and financial considerations ,also, earned the last rank. Considering this prioritization, the neural network used in this research contained data from 6 variables as an input variable, with two intermediate layers with 30 nodes in the first layer, and three nodes in the second layer,which had one outlet. The results indicated that the neural network model had the power to predict the success of the investment projects to1.2%of the error Manuscript profile -
Open Access Article
344 - The Modeling of Exchange Rate Predict in Iran by Using Neural Network Based on Genetic Algorithms and Particle Swarm Algorithm
ali jamali saeed daie karimzadehIn recent years the use of artificial intelligence techniques in the financial and investment markets instead of customary quantitative methods has been increasing and gives better performance towards classic methods usually. Artificial Neural Network (ANN), has weakn MoreIn recent years the use of artificial intelligence techniques in the financial and investment markets instead of customary quantitative methods has been increasing and gives better performance towards classic methods usually. Artificial Neural Network (ANN), has weaknesses points despite its enormous benefits also. In this study, in order to overcome the weaknesses of the network consists of combining artificial intelligence methods with Evolutionary algorithms, means of artificial neural network combined with genetic algorithm (GA) and Particle Swarm algorithm (PSO) to model and daily predict of nominal exchange rates or the exchange rate dollar by Rial in Iran in the period 21.03.2013 to 22.12.2019 is used. This combined model with neural networks method as one artificial intelligence model according to the criteria of MSE , RMSE, MAE, U.Theil compared. The results of this research show the superiority of synthetic neural network model -Particle Swarm algorithm compare to other models of investigation. Manuscript profile -
Open Access Article
345 - Presenting the developed model of benish model with emphasis on audit quality fea-tures using neural network, vector machine and random forest
Kiumars Pourgadimi Jamal Bahri Sales Saeed Jabbarzadeh Kangarluei Akbar Zavar RezaeePurpose: As the business process becomes more complex, the risk of financial statements being distorted increases with each passing day. In this regard, researchers have been looking for models to detect fraud in financial statements. The purpose of this research is to MorePurpose: As the business process becomes more complex, the risk of financial statements being distorted increases with each passing day. In this regard, researchers have been looking for models to detect fraud in financial statements. The purpose of this research is to present an expanded model based on the quality characteristics of the auditor.Methodology: Benish (1997) used a combination of financial ratios and accruals to predict profit manipulation methods. Since auditors are presented as external oversight in the corporate governance structure of the company's performance, in this study the model is developed based on the qualitative char-acteristics of the auditor, which include the auditor's size, auditor tenure, reporting delay, Auditor Class and Auditor Change. The fitting of the vector machine, random forest and neural network has been used to fit the ex-tended model.Findings: The results show that the coefficients obtained from the random forest model are 98.4% and more than the two neural network and vector model 93%. Also, the extended model is more accurate than the base model. Audit characteristics are influential in predicting fraud in financial statements and should be considered by capital market participants.Originality / Value: Research findings can be effective in improving the prediction of fraud in financial statements and also draw users' attention to the combination of financial statement information and the characteristics of the auditor's report in fraud prediction. Manuscript profile -
Open Access Article
346 - بهینه سازی تغییرات ریزساختاری و خواص مکانیکی آلیاژ آلومینیم A360 تولید شده به روش ریخته گری نیمه جامد
امین کلاه دوز محسن لوح موسویریخته­گری نیمه­جامدها، فرآیندی نوین می­باشد که می­تواند قطعاتی با ساختار یکنواخت کروی و با خواص مکانیکی بهبود یافته را تولید نمود. در این مقاله از روش مذاب­ریزی بر روی سطح شیب­دار برای تولید آلیاژ آلومینیم A360 استفاده شده است. با جریان یافتن مذا Moreریخته­گری نیمه­جامدها، فرآیندی نوین می­باشد که می­تواند قطعاتی با ساختار یکنواخت کروی و با خواص مکانیکی بهبود یافته را تولید نمود. در این مقاله از روش مذاب­ریزی بر روی سطح شیب­دار برای تولید آلیاژ آلومینیم A360 استفاده شده است. با جریان یافتن مذاب بر روی این سطح و اعمال تنش، فاز اولیه شاخه­ای در این آلیاژ به یک فاز غیرشاخه­ای تبدیل می­گردد. در این تحقیق تاثیر پارامترهایی از قبیل طول و زاویه سطح، دمای قالب و دمای بارریزی به عنوان متغیرهای فرآیند در نظر گرفته شده و مقادیر سختی و استحکام آلیاژ بر اساس تغییرات در پارامتر اندازه دانه مورد محاسبه قرار می­گیرد. در مرحله بعد، رابطه بین ورودی و خروجی­های فرآیند با استفاده از شبکه عصبی به دست می­آید. در نهایت رابطه به دست آمده با استفاده از الگوریتم ژنتیک بهینه می­شود. نتایج نشان می­دهد که با تغییر در دمای بارریزی و دمای قالب، سختی نمونه­ها نیز تغییر می­کند. به طوریکه با تغییر دمای بارریزی میزان 12% افزایش و با تغییر دمای قالب به میزان 5%، سختی افزایش می­یابد. همچنین طول و زاویه سطح شیب­دار نیز به ترتیب تاثیر 12% و 9% در مقدار افزایش سختی و 13% و 6% در میزان افزایش استحکام دارد. Manuscript profile -
Open Access Article
347 - بهینهسازی ابعاد پلیسه، قطعه کار و فاکتور اصطکاک در فورجینگ سرد با قالب بسته
مهدی ظهور حسین شاهوردی امین تفکریدر این مقاله، در طراحی فرایند فورجینگ با قالبهای بسته غیر دقیق برای قطعات متقارن محوری، سه پارامتر نسبت عرض به ضخامت پلیسه، نسبت ارتفاع به قطر قطعه کار (بیلت) و فاکتور اصطکاک، با روش اجزا محدود مورد بررسی قرار گرفته است. همچنین دو عامل مهم در فرایند فورج شامل نیروی لازم Moreدر این مقاله، در طراحی فرایند فورجینگ با قالبهای بسته غیر دقیق برای قطعات متقارن محوری، سه پارامتر نسبت عرض به ضخامت پلیسه، نسبت ارتفاع به قطر قطعه کار (بیلت) و فاکتور اصطکاک، با روش اجزا محدود مورد بررسی قرار گرفته است. همچنین دو عامل مهم در فرایند فورج شامل نیروی لازم برای فورج و مقدار پرشدن فضای داخلی قالب بررسی گردید. با کنترل ابعاد پلیسه، امکان کاهش مقدار دور ریز مواد مصرفی و افزایش درصد پرشدن قالب ایجاد میشود. برای اعتبار سنجی نتایج این کار تحقیقاتی، مقدار نیروی بهدست آمده از روش اجزا محدود با مقدار نیروی کار تجربی مقایسه شده است. برای هماهنگ کردن و ارتباط دادن بین پارامترهای مذکور و بهدست آوردن تابع عملکرد، شبکه عصبی دولایه بکار گرفته شده است. با به­کارگیری تابع به­دست آمده از شبکه عصبی و استفاده از الگوریتم ژنتیک، فاکتور اصطکاک و ابعاد پلیسه و قطعهکار برای کاهش مقدار نیرو و افزایش درصد پر شدن قالب بهدست آمد. سپس، این مقادیر با نتایج کار تجربی مرجع دیگر، مقایسه شده است. الگوریتم ژنتیک تخمین قابل قبولی برای پارامترهای مؤثر در فرایند فورج ارائه کرد، به طوریکه تطابق و نزدیکی خوبی بین نتایج این الگوریتم و روش تجربی وجود دارد. Manuscript profile -
Open Access Article
348 - بهینهسازی نرخ براده برداری در ماشینکاری تخلیه الکتریکی بر روی آلیاژ DIN 1.2080 با کمک روش شبکه عصبی و الگوریتم ژنتیک
مسعود عظیمی امین کلاه دوز سید علی افتخاریﻓﺮآﯾﻨﺪ ماشینکاری ﺗﺨﻠﯿﻪ اﻟﮑﺘﺮﯾﮑﯽ ﯾﮑﯽ از ﭘﺮﮐﺎرﺑﺮدﺗﺮﯾﻦ روشهای ماشینکاری غیر سنتی ﺑﺮای ﺑﺮاده ﺑﺮداری ﻗﻄﻌﺎت ﻫﺎدی ﺟﺮﯾﺎن اﻟﮑﺘﺮﯾﺴﯿﺘﻪ اﺳﺖ که دستیابی به قطعههایی باکیفیت سطح مناسب و نرخ برادهبرداری بالا در آن از اهمیت زیادی برخوردار است. با توجه به کاربرد وسیع و روزافزون Moreﻓﺮآﯾﻨﺪ ماشینکاری ﺗﺨﻠﯿﻪ اﻟﮑﺘﺮﯾﮑﯽ ﯾﮑﯽ از ﭘﺮﮐﺎرﺑﺮدﺗﺮﯾﻦ روشهای ماشینکاری غیر سنتی ﺑﺮای ﺑﺮاده ﺑﺮداری ﻗﻄﻌﺎت ﻫﺎدی ﺟﺮﯾﺎن اﻟﮑﺘﺮﯾﺴﯿﺘﻪ اﺳﺖ که دستیابی به قطعههایی باکیفیت سطح مناسب و نرخ برادهبرداری بالا در آن از اهمیت زیادی برخوردار است. با توجه به کاربرد وسیع و روزافزون آلیاژ DIN1.2080 در صنایع مختلف ازجمله قالبسازی، ابزارهای تراشکاری، برقوزنها، خانکشی، گیوتینهای برش و... بهدست آوردن شرایط بهینه ماشینکاری آن حائز اهمیت است. لذا هدف اصلی در این تحقیق ﺑﺮرﺳﯽ تأﺛﯿﺮ ﭘﺎراﻣﺘﺮﻫﺎی ورودی ﺷﺎﻣﻞ وﻟﺘﺎژ، شدتجریان، زﻣﺎن روﺷﻨﯽ ﭘﺎﻟﺲ و زمان خاموشی پالس ﺑﺮ روی نرخ برادهبرداری و بهینهسازی آن در ماشینکاری ﺗﺨﻠﯿﻪ اﻟﮑﺘﺮﯾﮑﯽ آلیاژ DIN1.2080 است لذا جهت دستیابی به نتیجه مطلوب پس از انجام آزمایشهای متعدد به کمک روش طراحی آزمایش تاگوچی و دترمینان بهینه بهمنظور پیشبینی و بهینهسازی نرخ برداشت براده از روش شبکه عصبی و الگوریتم ژنتیک استفادهشده است. در ادامه بهینهسازی پارامترهای ورودی بهمنظور بیشینه کردن نرخ برداشت براده صورت پذیرفت. در این حالت با فرایند کاهش زمان، کاهش هزینههای تولید بهدست میآید. پارامترهای بهینه در این آزمایش در شرایط شدتجریان 20 آمپر، ولتاژ 160 ولت، زمان روشنی پالس 100 میکروثانیه و زمان خاموشی پالس 12 میکروثانیه بهدست آمد که در این صورت به میزان نرخ برادهبرداری 063/0 سانتیمتر مکعب بر دقیقه دستیافته شد. سپس با انجام آزمایش صحهگذاری میزان خطا و دقت این روش سنجیده شد. با توجه به میزان خطای بهدست آمده که حدود 18/5 % بوده است روش استفاده شده برای الگوریتم ژنتیک مناسب ارزیابی شد Manuscript profile -
Open Access Article
349 - مدلسازی و مقایسه رفتار مکانیکی سازه های جدار نازک آلومینیومی با و بدون فوم پلی یورتان در نرم افزار LS-DYNA و ارائه یک مدل شبکه عصبی مصنوعی
میثم رستمی مجتبی حسنلو مصطفی سیاوشیقابلیت جذب انرژی سازه­های جدار نازک با مقاطع مختلف همواره مورد توجه محققان بوده است. این سازه­ها بعنوان جاذب­های انرژی در صنایع مختلف از جمله اتومبیل­سازی و هوافضا مورد استفاده وسیعی قرار می­گیرند و سبب حفاظت از سرنشینان و محوله­ها در حین برخورد Moreقابلیت جذب انرژی سازه­های جدار نازک با مقاطع مختلف همواره مورد توجه محققان بوده است. این سازه­ها بعنوان جاذب­های انرژی در صنایع مختلف از جمله اتومبیل­سازی و هوافضا مورد استفاده وسیعی قرار می­گیرند و سبب حفاظت از سرنشینان و محوله­ها در حین برخورد می­شوند. در این مقاله رفتار مکانیکی سازه­های جدار نازک از جنس آلومینیوم با فوم پرکننده پلی یورتان و بدون فوم تحت بارگذاری محوری ضربه­ای بررسی شده است. سازه­ها از نوع خیلی نازک می­باشند بطوریکه برای نمونه استوانه­ای رابطه (D/t) ≈ 550 برقرار است. تحلیل اجزا محدود و شبیه­سازی توسط نرم افزار LS-DYNA انجام شده است. سازه­های جدار نازک دارای مقاطع دایروی، شش ضلعی و چهار ضلعی با طول، ضخامت و اندازه محیط مقطع یکسان می­باشند. نتایج حاصل از پژوهش بیانگر آن است که سازه با مقطع دایروی از مقاطع چهار گوش و شش گوش انرژی بیشتری جذب می­نماید در حالیکه تغییر طول کمتری را تجربه می­کند. بعلاوه می­توان اثرات تمرکز تنش را در کنج­های مقاطع مربع و شش ضلعی بر روی جداره سازه­ها مشاهده کرد. همچنین سازه با مقطع دایروی به صورت متقارن­تر تحت بارگذاری دینامیکی فشرده می­گردد درحالیکه سازه­های جدار نازک با مقاطع شش و چهار ضلعی تمایل به کمانش دارند. در پایان نیز معماری از یک شبکه عصبی مصنوعی ارائه شده است تا با کمک آن و بهره­گیری از داده­های LS-DYNA بتوان رفتار جذب انرژی و نیروی این سازه­ها را در قالب مدلی در شبکه عصبی بیان نمود. نتایج مدل پیشنهادی در مقایسه با نتایج تحلیلی نرم­افزار LS-DYNA دقت قابل قبولی داشتند Manuscript profile -
Open Access Article
350 - Design and Explanation of Intelligent Adaptive Control Model in Strategic Planning with Financial Approach
zahra sadeghi Mohammad Reza Motadel abas toloiTo grow their business, organizations need accurate and up-to-date tools to develop sound financial planning strategies. The lack of comprehensive models that use modern tools and methods to facilitate strategic planning of organizations (with the least deviation from t MoreTo grow their business, organizations need accurate and up-to-date tools to develop sound financial planning strategies. The lack of comprehensive models that use modern tools and methods to facilitate strategic planning of organizations (with the least deviation from the strategic financial goals) is considered a research gap. In this study, a model to help formulate strategic planning of organizations in the financial sector (to make more accurate and precise decisions) presented. In this Intelligent Adaptive Control model, the Elman neural network has applied for system identification and adaptive algorithm process. The data used to build the model, according to the financial statements of Pars Oil Company between 1379 and 1397, have been obtained through Codal website. Profit-making indicators (including Operational Revenue, Other Operational Revenues & Costs, Other Non-Operational Revenues & Costs, Actual Costs, Executive & General Costs, Financial Costs, and Income Tax) used as a measure of a company's financial performance. The output of this model determines the allowed intervals of changes of these indicators to achieve the desired goal Manuscript profile -
Open Access Article
351 - Earthquake Risk Analysis in Azerbaijan Region Using Neural Network Method
Sayyed javad Sayyedfattahi Rouzbeh Dabiri Milad Farshbaf KhaliliIn this research, earthquake prediction and hazard risk analysis in Azerbaijan region have been done using neural network method. The forecasting approach is based on the use of artificial intelligence based on seismic data from previous times over a period of 100 years MoreIn this research, earthquake prediction and hazard risk analysis in Azerbaijan region have been done using neural network method. The forecasting approach is based on the use of artificial intelligence based on seismic data from previous times over a period of 100 years. Five characteristics of an earthquake event are extracted from data from previous years. The earthquake prediction model is based on five selected features with three different algorithms with feed forward neural network. Validation results indicate the high ability of the model to predict earthquakes in the study area. The results of the created models have been used to analyze the risk in the region of Azerbaijan. According to the obtained results, the region of Azerbaijan is prone to high earthquakes, which necessitates strict observance of standards in the construction of buildings. The results of the present study show that the Bayesian algorithm has the best performance in predicting earthquake risk. Manuscript profile -
Open Access Article
352 - Precipitation-runoff Simulation with Neural Network(Case study: Nasa Bam Plain)
mehdi shahrokhi sardoo mojtaba jafari kermanipourShort-term runoff forecasting is of particular importance due to its direct relationship with how managers interact with life risks caused by floods. In this research, by using artificial neural networks, simulation of rainfall-runoff process has been done on a daily ba MoreShort-term runoff forecasting is of particular importance due to its direct relationship with how managers interact with life risks caused by floods. In this research, by using artificial neural networks, simulation of rainfall-runoff process has been done on a daily basis in the Nasa Bam watershed. In order to predict the future process of using the water resources of the mentioned plain, different combinations of rainfall and temperature data and discharge and discharge difference of two consecutive days were used. The number of hidden layer neurons in the neural network varied between 2 and 10 neurons. The statistical criteria of root mean square error RMSE, mean absolute value of error MAE and correlation coefficient R were used to evaluate and compare the performance of neural networks in runoff forecasting. The results showed that by having 2 inputs and feedforward neural network or 1 input and newrbe network, the best performance was achieved and the rainfall-runoff process was predicted with higher accuracy. Manuscript profile -
Open Access Article
353 - ارائه مدلی برای پیش بینی سطح معنویت در سازمان های ورزشی با رویکرد شبکه عصبی مصنوعی
سید احسان امیرحسینی ابوذر زارع -
Open Access Article
354 - Prediction of Adhesion Parameters of Hook-shaped Steel Fibers and Concrete Using Artificial Neural Networks
Amir ebrahim akbari bagalAs steel fibers are important reinforcement materials in concrete, in this study, the behavior of hook-shaped steel fibers from concrete is predicted through the use of artificial neural networks. In the absence of comprehensive laboratory data, data obtained from finit MoreAs steel fibers are important reinforcement materials in concrete, in this study, the behavior of hook-shaped steel fibers from concrete is predicted through the use of artificial neural networks. In the absence of comprehensive laboratory data, data obtained from finite element analysis was used for modeling. The simulations are carried out using ABAQUS software's finite element method in 3D. Using the concept of the transition zone of the interface, whose parameters were obtained by inverse finite element analysis and experimental tests conducted on a sample of fibers, this model has been developed to simulate the interaction between fibers and concrete. On the basis of the results of the numerical model validated against the experimental results, the effective parameters of the fibers were extracted, and a neural network was then constructed based on the results. A multilayer forward perceptron artificial neural network and back-propagation training algorithm are used to predict pull-out force, with Marquardt-Lonberg optimization applied. The results demonstrate that the neural network model presented in this research is an effective method for predicting the pull-out force of fibers from concrete, in part because it allows the use of more variables in modeling, as well as delivering more accurate results. Manuscript profile -
Open Access Article
355 - Comparison of Artificial Neural Network and Regression Methods in Predicting the Modulus of Deformation of Stone using Dilatometry Test.
Manouchehr Hoseine Rouzbeh Dabiri Larissa KhodadadiIn geotechnical engineering, the modulus of deformation (Em) is actually the ratio of stress to strain. The application of this module is in the fields of dam construction, tunnel construction, road construction, etc. Today, there are various methods to obtain the defor MoreIn geotechnical engineering, the modulus of deformation (Em) is actually the ratio of stress to strain. The application of this module is in the fields of dam construction, tunnel construction, road construction, etc. Today, there are various methods to obtain the deformation modulus, among which we can refer to in-situ tests (loading plate-dilatometry), laboratory tests, and practical relationships. Also, there are different methods to predict and determine the relationships between several different parameters, which can be referred to regression analysis and artificial neural network. The main goal of the present research is to provide a new relationship to predict the modulus of deformation of rocks before performing the dilatometry test with the least error. The results of the studies have shown that neural network modeling is more efficient than regression analysis in all input independent variables, and it has a higher level of confidence only with the input of Q parameter to the regression analysis equation. Also, by comparing these two methods, it was found that the more the number of input variables, the better the neural network works. Manuscript profile -
Open Access Article
356 - Comparison of forecasting ability of artificial neural network with other forecasting methods: case of sugar beet price
Hamid Mohammadi Farshid Kafilzadeh Mohammad Naghshinehfard Siyamak PishbinThe aim of this study was to forecast nominal and real price of sugar beet and to compare forecasting ability of artificial neural network method with other forecasting methods. The stationary of the series was tested and then, in order to investigate whether series are MoreThe aim of this study was to forecast nominal and real price of sugar beet and to compare forecasting ability of artificial neural network method with other forecasting methods. The stationary of the series was tested and then, in order to investigate whether series are stochastic, nonparametric test of Vald-Wulfowitz and parametric test of Durbin-Watson were applied. Based on the above tests results, nominal price of sugar beet were recognized non-stochastic and predictable, while the real price series was found stochastic. The study period covers 1971-2005. The models used for forecasting were autoregressive, moving average, ARIMA, Single and Double exponential smoothing, harmonic, ARCH and artificial neural network. Based on the lowest forecasting error criterion, harmonic model forecasted nominal price of sugar beet with lowest forecasting error. The amount of nominal series forecasted by different models was at range of 344000-396000 and 398000-448504 rials per ton for 2004 and 2005, respectively. The happened values of nominal price series for 2004 and 2005 were 387200 and 447000 rials per ton, respectively. Manuscript profile
Sanad
Sanad is a platform for managing Azad University publications
Links
Related Centers
Technical Support
Official pages
The rights to this website are owned by the Raimag Press Management System.
Copyright © 2021-2024