Comparing the Performance of Machine Learning Techniques in Detecting Financial Frauds
محورهای موضوعی : Financial Economics
Jafar Nahari Aghdam Qala Jougha
1
,
Nader Rezaei
2
,
Yagoob Aghdam Mazraee,
3
,
Rasol Abdi
4
1 - Department of Accounting, Bonab branch, Islamic Azad University, Bonab, Iran.
2 - Department of Management and Accounting, Bonab Branch, Islamic Azad University, Bonab, Iran
3 - Department of Accounting,sofian,branch,islamic azad university,sofian,iran
4 - Department of Accounting, Bonab Branch, Islamic Azad university, Bonab, Iran
کلید واژه: Bayesian Linear Regression, Neural Network, Logistic Regression, Financial Fraud,
چکیده مقاله :
Detecting financial fraud is an important process in the activities of compa-nies. In the last decade, much attention has been paid to fraud detection techniques. Financial fraud is a problem with far-reaching implications for shareholders. Today, financial fraud in companies has become a big prob-lem. Companies and regulatory agencies must continuously develop their mechanisms to detect fraud. Machine learning and data mining techniques are currently commonly used to solve this problem. However, these tech-niques still need to be improved in terms of computational cost, memory cost, and dealing with big data that is becoming a feature of current financial transactions. In this research, machine learning techniques including logistic regression, neural network, and Bayesian linear regression were used to de-tect financial frauds in the Iranian stock market. According to the obtained results, the support vector machine model with radial kernel has the lowest RMSE and the highest accuracy criterion, and the support vector machine model with linear kernel and Bayesian linear regression has the highest RMSE and the lowest accuracy criterion for modeling the financial fraud of companies in they were Tehran stock market. Also, the models of artificial neural network model, Bayesian linear regression and support vector ma-chine model with linear kernel respectively had the lowest characteristic values and did not perform relatively well in detecting the existence of fi-nancial fraud in the companies present in the Tehran stock market.
Detecting financial fraud is an important process in the activities of compa-nies. In the last decade, much attention has been paid to fraud detection techniques. Financial fraud is a problem with far-reaching implications for shareholders. Today, financial fraud in companies has become a big prob-lem. Companies and regulatory agencies must continuously develop their mechanisms to detect fraud. Machine learning and data mining techniques are currently commonly used to solve this problem. However, these tech-niques still need to be improved in terms of computational cost, memory cost, and dealing with big data that is becoming a feature of current financial transactions. In this research, machine learning techniques including logistic regression, neural network, and Bayesian linear regression were used to de-tect financial frauds in the Iranian stock market. According to the obtained results, the support vector machine model with radial kernel has the lowest RMSE and the highest accuracy criterion, and the support vector machine model with linear kernel and Bayesian linear regression has the highest RMSE and the lowest accuracy criterion for modeling the financial fraud of companies in they were Tehran stock market. Also, the models of artificial neural network model, Bayesian linear regression and support vector ma-chine model with linear kernel respectively had the lowest characteristic values and did not perform relatively well in detecting the existence of fi-nancial fraud in the companies present in the Tehran stock market.
[1] Keshavarz, Ahmad, Ghasemian, Hassan, A Fast Algorithm Based on Support Vector Machine for Classifying Hyperspectral Images Using Spatial Correlation, Iranian Electrical and Computer Engineering Journal, 2014; 37-44.
[2] Karanjadi, Ayding, Pourqasmi, Hamidreza , Landslide susceptibility assessment using data mining models, a study Case: Chahalchai Watershed, 2018; 11(1) : 42
[3] Kim, K., Financial Time Series Forecasting Using Support Vector Machines, Neurocomputing 2003;55:307-319.
[4] Elith, J., J.R. Leathwick and T. Hastie, a working guide to boosted regression trees, Journal of Animal Ecology, 2008; 77(4): 802-813.
[5] Kamrani, Hossein, & Abedini, Bijan, Formulation of financial statement fraud detection model using artificial neural network and support vector machine methods in companies admitted to Tehran Bahadur Stock Exchange, Knowledge of Accounting and Management Audit, 1401;11(41):285-314.
[6] Abeare, S.M., Comparisons of boosted regression tree, GLM and game performance in the standardization of yellowfin tuna catch-rate data from the Gulf of Mexico online fishery. 2009
[7] Moradi, Mohsenshafiei Sardasht, Morteziabrahimpour, Maleeha, Predicting the financial distress of companies by support vector machine models and multiple audit analysis, Stock Exchange Quarterly, 2013; No. 18, 5(9): 113-136.
[8] Broghni Mehdi, Porfashmi Sima, Zarei Mehdi, Aliabadi Kazem, Spatial modeling of the sensitivity of dust centers to its emission in eastern Iran using BRT enhanced regression tree model, Geographical studies of dry areas. 1398; 9 (35): 14-28
[9] Kamrani, Hossein, & Abedini, Bijan, Formulation of financial statement fraud detection model using artificial neural network and support vector machine methods in companies admitted to Tehran Bahadur Stock Exchange, Knowledge of Accounting and Management Audit, 1401;11(41): 285-314.
[10] Essani, Elahe, Sepasi, Dr. Sahar, Etemadi, Dr. Hossein, Azar, Dr. Adel, presenting a new approach in predicting and detecting financial statement fraud using the bee algorithm, Journal of Accounting Knowledge, (2018; 10(3): 139-167. doi: 10.22103/jak.2019.13616.2927
[11] Che. Tsai, Hung. C, Automatically Annotating Images With Keywords: A Review Of Image Annotation Systems, Recent Patents On Computer Science, 2008; 55-68.
[12] Camps-Valls, G., Tuia, D., Gomez-Chova, L., Jimenez S. and Malo, J., Remote Sensing Image Processing, Morgan & Claypool Publishers, 2012;176
[13] Yara Alghofaili, Albatul Albattah & Murad A. Rassam, A Financial Fraud Detection Model Based on LSTM Deep Learning Technique, Journal of Applied Security Research, 2020; doi: 10.1080/19361610.2020.1815491
