Hierarchical clustering-based suspicious taxpayer detection method

Abstract

The invention discloses a hierarchical clustering-based suspicious taxpayer detection method. The hierarchical clustering-based suspicious taxpayer detection method includes steps of 1) according to interest relationship between the taxpayers, performing clustering based on a hierarchical clustering algorithm, and according to association degrees between the taxpayers, dividing the taxpayers into clusters which are interest associated societies; 2) with analysis of marked taxpayer sample data, extracting distinction features between the normal taxpayers and the abnormal taxpayers based on complex network indicators, such as degree distribution and clustering coefficient; and 3) calculating feature similarity degree between identified interest associated societies and the normal taxpayers, as well as between the identified interest associated societies and the abnormal taxpayers, and further detecting the suspicious taxpayers.

Description

technical field [0001] The invention belongs to the subfield of data mining in computer science and technology, and relates to a method for detecting suspicious taxpayers, in particular to a method for detecting suspicious taxpayers based on hierarchical clustering. Background technique [0002] In the tax system in the real environment, the information between the tax systems in different regions is asymmetric, the information is closed and not unified, it is difficult to truly count the tax data of each region, a large amount of tax data lacks analysis and utilization, and it is difficult to find a large number of tax evasion and fraud. As a result, the loss rate of tax revenue and the loss of tax money remain high. It is very important to analyze the tax data, extract interest-related associations through the interest association between taxpayers, and finally discover and combat tax evasion, and become the focus and difficulty of the tax analysis system, but none of the ...

AI Technical Summary

Problems solved by technology

[0002] In the tax system in the real environment, the information between the tax systems in different regions is asymmetric, the information is closed and not unified, it is difficult to truly count the tax data of each region, a large amount of tax data lacks analysis and utilization, and it is difficult to find a large number of tax evasion and fraud. Leading to high tax leakage rate and tax leakage

It is very important to analyze the tax data, extract interest-related associations through the interest association between taxpayers, and finally discover and combat tax evasion, and it has become the focus and difficulty of the tax analysis system, but none of the existing technologies can be effective. Judging whether there are suspicious taxpayers in the interest-related associations, so the behavior of discovering and cracking down on tax evasion and tax evasion becomes quite severe

Images