Data Mining Techniques Applied in Tax Administrations: A Literature Review

Because the tax administrations have the control of evasion and the risk of non-payment among their main activities, they have used historical information and analysis techniques that allow to identify the source of the evasion problem, discover information, select high-risk taxpayers, create models, offer new services, design methods to control taxpayers, and evaluate decisions. However, the tax administrations does not have unlimited resources for control process. Consequently, the management model used in tax administrations must seek the voluntary compliance of taxpayers through intelligent controls with efficient use of resources. Additionally, those controls must explore a large amount of structured, semi-structured and unstructured information from different sources of information that has been accumulated over the years. This study aims to present a literature review about the use of data mining techniques in tax administrations. Therefore, we analyze works in scientific data bases and documents of organizations that support tax administrations. Our study focuses on identifying problems that could be resolved with data mining, limits that have been risen, results that have been obtained by applying data mining in tax administrations and techniques that have been used. Our study could be useful to devise how data mining could support the activities of tax administrations and encourage the application of new data mining techniques in the analysis of information.