IOSR Journal of Computer Engineering (IOSR-JCE) e-ISSN: 2278-0661,p-ISSN: 2278-8727 PP 05-08 www.iosrjournals.org International conference on computing and virtualization (ICCCV-17) 5 | Page Thakur college of Engineering and Technology More Focus on Tax Evasion Detection with Graph Based Approach Komal Rokade 1 , Rashmi Mane 2 1 (CST, Department of technology, Shivaji University, India) 2 (CST, Department of technology, Shivaji University, India) Abstract: A tax is the source of government funding. The purpose of tax is to increase revenue to fund government. The money paid by taxpayers in taxes goes to many places. It is used to paying the salaries of government workers, tax money also help to support common resources, such as police and firefighters. Tax money helps to ensure the roads you travel on are safe and well-maintained. Taxes fund public libraries and parks. Tax evasion is increased so tax evasion detection is very important in current status to avoid loss of government funding. Taxpayers are required to store and update, on an annual basis, a set of documents and information relating to international transactions or specified domestic transactions. In recent work on tax evasion detection is done but it is not effective some drawbacks are there. This article gives an introduction to related work done in tax evasion detection and describes the methods of tax evasion. Auditing is very important to find out tax evasion, and data mining techniques are applied to select record for audit, also data mining techniques are applied in tax evasion detection. Keywords: Group mining, Tax evasion, TPIIN, trading data I. Introduction Introduction The main reason that data mining has attracted a great deal of attention in information industry in recent years is due to the wide availability of huge amounts of data and the need for turning such data into useful information and knowledge. Data mining is nothing but the extracting or mining knowledge from large amounts of data. Many people uses data mining as an alternate for term, Knowledge Discovery in Databases", or KDD. Alternatively, data mining is essential and important step in the process of knowledge discovery in databases. Tax evasion and tax fraud have been a constant issue for tax administrations, especially when pertaining to developing countries. While it is true that taxes are the source of government earning, the reality is that they send a very important signal about the commitment and effectiveness with which the State can carry out its functions and restrict access to other sources of income. Tax evasion is illegal evasion of taxes by individuals and corporations. The number of annual tax related business records is up to 1 billion, the daily peak of these records is up to ten million. This volume of data challenges traditional data mining based methods of tax evasion. The results of the clustering based and neural network based methods are not explainable and their tax evasion identification efficiency is low. When talking about the properties of big data, traditional data mining-based methods have their limitations. The classification-based methods need a set of sample data for training, which means the data need to be manually labeled before training takes place. Moreover, the trained model is sensitive to the sample data and will be out- of-date if behaviors in tax evasion change. In addition, the results derived from clustering-based methods and neural network-based methods are difficult to explain and trace. The worse thing is that the above mentioned data-mining-based methods need to search and evaluate each transaction in the tax-oriented big data before reliable outcomes can be derived. The proposed method is more effective and efficient than the existing approaches, as it aims to select the suspicious relations first via other related data sources and then identify those suspicious transactions. II. Related Work “Destination taxation and evasion: Evidence from U.S. inter-state commodity flows” in 2014 by W. F. Fox, L. Lunab, and G. Schau,[1] in that they developed a new way to examine tax evasion that focuses on corporate transactions, rather than corporate profits. Specifically, they examine how commodity flows respond to destination sales taxes, allowing for tax evasion as a function of distance between trade partners. After accounting for transportation costs, they find that the effect of taxes decreases as distance increases. Due to this way longer distances between trade partners avoid smooth government supervision and increase the