A Survey of Tax Risk Detection Using Data Mining Techniques

• DOI: 10.1016/j.eng.2023.07.014
• Publication Date: 2023-09-25T22:10:27Z
• Authors: Qinghua Zheng; Yiming Xu; Huixiang Liu; Bin Shi; Jiaxiang Wang; Bo Dong
• Keywords: Intellectualization; Knowledge guide; 0202 electrical engineering, electronic engineering, information engineering; Informatization; 02 engineering and technology; TA1-2040; Tax risk detection; Engineering (General). Civil engineering (General); Data mining

Tax risk behavior causes serious loss of fiscal revenue, damages the country's public infrastructure, and disturbs market economic order fair competition. In recent years, tax detection, driven by information technology such as data mining artificial intelligence, has received extensive attention. To promote high-quality development detection methods, this paper provides first comprehensive overview summary existing methods worldwide. More specifically, it discusses negative impacts behavior, along with detection. It then focuses on data-mining-based utilized around world. Based different principles employed algorithms, can be divided into two categories: relationship-based non-relationship-based. A total 14 are identified, each method is thoroughly explored analyzed. Finally, four major technical bottlenecks current data-driven analyzed discussed, including difficulty integrating using fragmented knowledge, unexplainable results, high cost reliance algorithms labeled information. After investigating these issues, concluded that knowledge-guided big knowledge engineering will trend in field future; is, gradual transition from informatization to intelligence future direction.