Examining the Role of Machine Learning Algorithms in Identifying Tax Evasion among Taxpayers of the Tehran Provincial Tax Administration

Abstract

Tax evasion is considered one of the most significant challenges facing economic and taxation systems in developing countries. It leads to a reduction in government revenue, disruption of tax equity, and a decline in economic transparency. In recent years, advances in information technology and the emergence of artificial intelligence algorithms, particularly machine learning, have created suitable conditions for big data analysis and the identification of hidden patterns in taxpayers’ financial behavior. The present study aims to examine the role of machine learning algorithms in identifying tax evasion among taxpayers of the Tehran Provincial Tax Administration. This research is applied in terms of purpose and descriptive-analytical in terms of methodology. The statistical population includes the tax records of individual and corporate taxpayers registered with the Tehran Provincial Tax Administration during the years 2019 to 2023. In this study, the ID3 decision tree and Bayesian algorithms were used to analyze the data. The expected results indicate that machine learning algorithms have higher accuracy than traditional methods in identifying high-risk taxpayers and predicting tax evasion. The findings of this study can contribute to improving the tax auditing process, reducing monitoring costs, and increasing government tax revenues.