There is a new tendency for corporations to evade tax via Interest Affiliated Transactions (IAT) that are controlled by a potential ‘Guanxi’ between the corporations’ controllers. At the same time, the taxation information-related data is a classic kind of big data. The issues challenge the effectiveness of traditional data mining-based tax evasion detection methods. To address this problem, this chapter first proposes a colored and weighted network-based model (CWNBM) for characterizing economic behaviors, social relationships and the IATs between corporations and generating a heterogeneous information network – Corporate Governance Network (CGN). Next, a definition of controller interlock is coined, which characterizes the interlocking relationship between corporations’ controllers. Then, to accomplish the task of detecting controller interlock-based tax evasion groups in a corporate governance network, (i) graph projection method for recognizing controller interlock pattern and (ii) component pattern matching method for detecting suspicious groups are introduced. Experimental results, based on seven-year period, 2009–2015, of one province in China, demonstrated that our proposed method can greatly improve the efficiency of tax-evasion detection.
|Otsikko||Machine Learning for Computer and Cyber Security|
|Alaotsikko||Principle, Algorithms, and Practices|
|Toimittajat||Brij B. Gupta, Michael Sheng|
|Tila||Julkaistu - 7 helmikuuta 2019|
|OKM-julkaisutyyppi||A3 Kirjan osa tai toinen tutkimuskirja|