Comparing the Performance of Machine Learning Techniques in Detecting Financial Frauds

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.