Using data mining techniques to measure tax risk of value added taxes

Subject Areas : Journal of Investment Knowledge

Mohammad Masihi ¹ , Ahmad Yaghoobnejad ² , Amirreza Keyghobadi ³ , Taghi Torabi ⁴

1 - Ph.D student, financial management department, UAE branch , islamic azad universityt
2 - Associate professor of accounting department, centeral branch of tehran, islamic azad university
3 - Asistant professor of Department of Accounting , centeral branch of tehran, islamic azad university
4 - associat profesoor of economics, science and research branch of tehran, islamic azad universit

Received: 2019-01-02 Accepted : 2019-04-14 Published : 2019-12-22

Keywords: data mining techniques, tax risk, value added taxes, tax audit,

Abstract :

In this paper using data mining to studied taxpayers risk value added taxes. the importance of assessing the taxpayers risk of value added taxes in order to formulate an effective plan for choosing taxpayers for tax audit with the goal of increasing efficiency and effectiveness, in the country's value added taxes system. In this research taxpayers are catogorized into three, risk_free , low_ risk and risk _averse groups. To assess tax risk two techniques, data mining machin backup vector and logistic regression have been used. The research community consist of large legal entities in Tehran.that wich have been subject to tax audit in value added taxes system in 2012 to 2015. In this research, variables are include corporate governance mechanisms, special corporate features, the nature of the activity of the pioneers of the control system and tax ratios wich are used to train and use the model. The research's results show two techniques LSVM ,Logistic, have a reliability of 70percent and a kind of integration into the results of these two techniques has been achieved nearly 83 percent of reliability has a higher potential.

References:

*       موسوی جهرمی، یگانه، فرهاد طهماسبی بلداجی و نرگس خاکی (1388)، «فرار مالیاتی در نظام مالیات بر ارزش افزوده: یک مدل نظری»، فصلنامه تخصصی مالیات، دوره جدید، شماره پنجم (مسلسل 53)، ص27-38.

*       پیری، صفر و جباری، حسین (1394) " عوامل مرتبط با به کارگیری حسابرسی داخلی مبتنی بر یسک"، اولین همایش  ملی حسابداری و حسابرسی

*       رحمانی و ابوحمزه (1394) "حسابرسی مالیاتی و تمکین مالیات شرکتی"، مجله حسابرسی، سال اول، شماره 2 ، پاییز 93

*       دستگیر، محسن و غریبی، مریم (1394) « کاربست روش‌های داده‌کاوی به منظور ارتقای‌ عملکرد و تشخیص فرار مالیاتی» پژوهشنامه مالیات، ش 208 ص 116ـ95

*     رمضانی، سید مهدی و دستگیر، محسن و عسکری، علی و خانی، عبدالله (1392)، «حسابرسی مالیاتی مبتنی بر ریسک و معیارهای تعیین کننده آن با تأکید بر اندازه مؤدیان مالیاتی و تجربه سایر کشورها». پژوهش نامه مالیات، شماره 66، تابستان 1392.

*       سازمان مالیاتی ایران (شهریور1391)، «ظرفیت‌های مالیاتی اقتصاد ایران، چالش‌ها و راه‌کارهای توسعه آن»، دفتر پژوهش و برنامه ریزی.

*       Alabede J O, Zaimah B t. Zainol Ariffin and Kamil Md Idris.2011.Determinants of Tax compliance Behaviour : A Proposed Model for Nigeria.International Rsearch Journal of  Finance and  Economics,ISSN1450-2887 Issue78.

*       Alabede J O, Zaimah Zainol Ariffin and Kamil Md Idris.2011.Individual Taxpayers Attitude and compliance Behaviour in Nigeria: The Moderating Role of  Financial Condition and Risk Preference. Journal of Accounting  and  Taxation Vol.3(5) ,pp.91-104

*       Abebaw Kassie, Melat, Abera (2016), Factors Affecting Tax Audit Effectiveness Evidence From Large Tax Payers Office of Ethiopian Revenue And Customs Authority , Thesis - Accounting & Finance

*       James Alm, Michele Bernasconi , Susan Laury , Daniel J. Lee (2016), Culture, Compliance, and Confidentiality: Taxpayer Behavior in the United States and Italy, University Ca' Foscari of Venice, Dept. of Economics Research Paper Series No. 36

*        Kuo-Wei Hsu , Nishith Pathak, Jaideep Srivastava, Greg Tschida, Eric Bjorklund (2015), “Data Mining Based Tax Audit Selection: A Case Study of a Pilot Project at the Minnesota Department of Revenue”, Real World Data Mining Applications, Volume 17 of the series Annals of Information Systems pp 221-245

*       20. Khwaja, Munawer Sultan, Rajul Awasthi, and Jan Loeprick (2011). "Risk-Based Tax Audits: Approaches and Country Experiences", The International Bank for Reconstruction and Development / The World Bank, 1818 H Street NW, Washington DC20433.

*       LIDIJA HAUPTMAN, MIRJANA HORVAT & ROMANA KOREZ-VIDE (2014) Improving Tax Administration's Services as a Factor of Tax Compilance: The Case of Tax Audit, LEX LOCALIS - JOURNAL OF LOCAL SELF-GOVERNMENT Vol. 12, No. 3, pp. 481-501, July 2014

*       Lu, S.X. and Wang, X.Z., 2004, August. A comparison among four SVM classifaction methods:LSVM, NLSVM. SSVM and NSVM. In Machine learning and cybernetics,2004. Proceeding of 2004 international conference on (vol.7,pp.4277-4282). Ieee.

*       Rayisannkar, P., V. Ravi, G. Raghava Rao & Bose. (2011). Detection of Financial Statement Fraud and Feature Selection Using Data Mining Techniques. Decision Support Systems, 50,491-500