Journal of Advances in Computer Research: 5 (4)

Open Access Article

1 - A Novel Approach to Feature Selection Using PageRank algorithm for Web Page Classification
Farhad Rezvani Farhad Soleimanian Gharehchopogh

In this paper, a novel filter-based approach is proposed using the PageRank algorithm to select the optimal subset of features as well as to compute their weights for web page classification. To evaluate the proposed approach multiple experiments are performed using acc More

In this paper, a novel filter-based approach is proposed using the PageRank algorithm to select the optimal subset of features as well as to compute their weights for web page classification. To evaluate the proposed approach multiple experiments are performed using accuracy score as the main criterion on four different datasets, namely WebKB, Reuters-R8, Reuters-R52, and 20NewsGroups. By analyzing the obtained results, it is observed that the accuracy score of the classifier on WebKB, Reuters-R8, and Reuters-R52 datasets significantly improved from 91% up to 96% compared to the best result achieved by other feature selection methods like IG and Chi-2. Whereas, the accuracy score of the classifier on 20NewsGroups dataset didn't see any noticeable improvement and remained close to the most compared methods. Evaluating the performance of the proposed approach shows the superiority of it in obtaining higher accuracy scores when compared with the feature sets selected by other methods. Manuscript profile

Open Access Article

2 - Simultaneous Classification and Traction of Moving Obstacles by LIDAR And Camera Using Bayesian Algorithm
Masrour Dowlatabadi Ahmad Afshar Ali Moarefianpour

Shortly, preventing collisions with fixed or moving, alive, and inanimate obstacles will appear to be a severe challenge due to the increased use of Unmanned Ground Vehicles (UGVs). Light Detection and Ranging (LIDAR) sensors and cameras are usually used in UGV to detec More

Shortly, preventing collisions with fixed or moving, alive, and inanimate obstacles will appear to be a severe challenge due to the increased use of Unmanned Ground Vehicles (UGVs). Light Detection and Ranging (LIDAR) sensors and cameras are usually used in UGV to detect obstacles. The tracing and classification of moving obstacles is a significant dimension in developed driver assistance systems. The present study indicated a multi-hypotheses monitoring and classifying approach, which allows solving ambiguities rising with the last methods of associating and classifying targets and tracks in a highly volatile vehicular situation. We proposed a recursive method based on Bayesian Algorithm for using classification information of obstacles in the tracking information of them and vice versa. The results are shown that the proposed method can improve classifying and tracking together.This method was tested through real data from various driving scenarios and focusing on two obstacles of interest vehicle and pedestrian. Manuscript profile

Open Access Article

3 - Design and Analysis of a Fault Tolerant 3-Input Majority Gate in Quantum-dot Cellular Automata
Somayyeh Jafarali Jassbi Farzaneh Jahanshahi Javaran Hossein Khademolhosseini Amir Sabbagh Molahosseini

QCA is a kind of computational technology used for developing circuits in Nano sizes. Decreasing the dimensions of pieces has led to the increase of circuit sensibility and quantum circuits are more vulnerable to defects and radiations of the environment. Majority gate More

QCA is a kind of computational technology used for developing circuits in Nano sizes. Decreasing the dimensions of pieces has led to the increase of circuit sensibility and quantum circuits are more vulnerable to defects and radiations of the environment. Majority gate and NOT gate (inverter) are the two basic gates in QCA technology based on which almost all circuits are made. So far, a limited number of fault-tolerant majority gates have been presented and research in this particular field seems appropriate. In this research we intend to provide a comprehensive design of 3-input majority gate in quantum cellular automata for all possible faults: misalignment, missing, dislocation, and redundancy so that low overhead is added to circuit. The gate is made up of both 90-degree and 45-degree cells. The results of this study indicate that our proposed 3-input majority gate is more fault-tolerant to the defects compared to the formerly presented one. Manuscript profile

Open Access Article

4 - NMFA: Novel Modified FA algorithm Based On Firefly Recent Behaviors
Fatemeh Jafarnejad Rezaiyeh Kambiz Majidzadeh

The Firefly optimization algorithm (FA) is one of the practical nature-inspired metaheuristic approaches in 2008, which simulated the behavior of fireflies in the movement toward the light sources. Recent studies on this beautiful creature have revealed new behaviors th More

The Firefly optimization algorithm (FA) is one of the practical nature-inspired metaheuristic approaches in 2008, which simulated the behavior of fireflies in the movement toward the light sources. Recent studies on this beautiful creature have revealed new behaviors that strongly require us to review them. The proposed algorithm NMFA is the simulation results with the latest information from the behavior of fireflies. The NMFA is used for data clustering and optimization of continuous problems. The experimental results of the testing on optimization of 26 standard functions show that the proposed method works best in terms of success rate and convergence than the FA, HS, ABC, and IWO algorithms and makes an important and substantial difference in optimization. The non-parametric, statistical, and pairwise tests show the superiority of the modern firefly algorithm. The NMFA can cluster the datasets like the conventional K-means algorithm and obtain a significant result among the well-known methods. Manuscript profile

Open Access Article

5 - MCSM-DEEP: A Multi-Class Soft-Max Deep Learning Classifier for Image Recognition
Aref Safari

Convolutional neural networks show outstanding performance in many image processing problems, such as image recognition, object detection and image segmentation. Semantic segmentation is a very challenging task that requires recognizing, understanding what's in the imag More

Convolutional neural networks show outstanding performance in many image processing problems, such as image recognition, object detection and image segmentation. Semantic segmentation is a very challenging task that requires recognizing, understanding what's in the image in pixel level. The goal of this research is to develop on the known mathematical properties of the soft-max function and demonstrate how they can be exploited to conclude the convergence of learning algorithm in a simple application of image recognition in supervised learning. So, we utilize results from convex analysis theory which associated with hierarchical architecture to derive additional properties of the soft-max function not yet covered in the existing literature for Multi-Class Classification problems. The proposed MC-DEEP model represents an average accuracy of 90.25% in different layers setting with 95% confidence interval in best initial settings in deep convolutional layers which applied on MNIST dataset. The results show that the regularized networks not only could provide better segmentation results with regularization effect than the original ones but also have certain robustness to noise. Manuscript profile

Open Access Article

6 - Sentimental Categorization of Persian News Headlines using Three Machine Learning Techniques Versus Human Categorization
Vahid Mirzaeian

The aim of this paper is to elaborate on an attempt to classify Persian news headlines using machine learning techniques rather than human-based analysis. Three major techniques namely Naïve Bayes, Maximum Entropy and Support Vector Machine were introduced and appl More

The aim of this paper is to elaborate on an attempt to classify Persian news headlines using machine learning techniques rather than human-based analysis. Three major techniques namely Naïve Bayes, Maximum Entropy and Support Vector Machine were introduced and applied to Persian news headlines. Results were compared with each other as well as the human analysis. It is concluded that these techniques outperform human analysis and one technique (Naïve Bayes) is superior to all the techniques mentioned. It can be concluded from this study that the inclusion of discourse analysis is necessary in order to attain better results since the whole is not necessarily the sum of the parts. It means that what you see in the headline does not necessarily reflect what is mentioned in the news itself. So it is recommended that in future studies, elements from discourse analysis be introduced into these algorithms so that better results can be achieved. Manuscript profile

Open Access Article

7 - Face recognition using sparce reprasentations and p-laplacian
Homayun Motameni

Face recognition is one of the most important identification tools in biometrics. Nowadays, the topic of face recognition has many applications in various fields, including public security, identity identification, protection of important and sensitive places, access co More

Face recognition is one of the most important identification tools in biometrics. Nowadays, the topic of face recognition has many applications in various fields, including public security, identity identification, protection of important and sensitive places, access control, video surveillance, and so on. Two important issues in face recognition applications are speed and accuracy in detection. Various studies have shown that face recognition through Sparce Representation Classification (SRC) works very well.The purpose of this paper is to propose a fast and efficient method for the sparse representation-based face recognition .Due to the fact that retrieving the Sparce Representation based on L1 norm optimization for a large dictionary has a large computational volume, a Smooth L0 norm optimization (SL0) method is used. Also, due to the fact that one of the challenges of face recognition is the existence of brightness changes in images, so we use the P- Laplacein algorithm in the feature extraction step to give us more complete information about the face image by recognizing the edge. As the simulation results on the Extended Yale B and AR database show, the proposed hybrid method has a higher detection rate than the sparse display method. Manuscript profile

Open Access Article

8 - A New Replica Consistency Maintenance Method Based on Dynamic Update Rate Calculation in the Data Grid
Rasmieh Parvaneh Mahsa Beigrezaei

Data Grid provides convenient services for data management and uniform access to distributed high volume data sources. Data replication is a service to facilitate and accelerate data accessing in Data Grid. But in the large-scale grid environment, the need to maintain r More

Data Grid provides convenient services for data management and uniform access to distributed high volume data sources. Data replication is a service to facilitate and accelerate data accessing in Data Grid. But in the large-scale grid environment, the need to maintain replica consistently and ensure identical content of the various replica of a file are inevitable. This paper presents an effective consistency method that maintains replicated data consistent. In this method, the node which contains replica is responsible for managing consistency. In each node, the replica update rate is calculated based on the rate of change in the original data file and the rate of the access request to the existing replica version. The proposed model reduces access delays. The model was simulated using OptorSim developed by European Data Grid projects. The experimental results show that our proposed model has better performance in comparison with optimistic and pessimistic approaches in terms of job execution time, effective network usage, and the total number of updates. Manuscript profile

XML

List of Articles