TaxThemis: Interactive Mining and Exploration of Suspicious Tax Evasion Group

Yating Lin, Kamkwai Wong, Yong Wang, Rong Zhang, Bo Dong, Huamin Qu, Qinghua Zheng,

View presentation: 2020-10-30T16:00:00Z GMT-0600 Change your timezone on the schedule page
Exemplar figure
The system interface of TaxThemis: (A) The Control Panel consists of a bar chart to show the temporal summary of the daily related party's daily transaction amounts, together with a parameter selector to support network fusion through interactively setting the preferred period or relevant thresholds. (B) The Group Overview visualizes the topology of related party transactions in each suspicious groups and ranks the groups by their group features. This helps users focus on the most suspicious groups. (C) The Graph View shows the hierarchical investment relationships and related party transactions within the selected suspicious group. It supports group assessment in revealing the common beneficial owners and former tax evaders. (D) The Detail View presents the profit status of the traders who conducted the selected related party transactions, revealing their suspicious behavior of transferring revenue transfers
Fast forward

Direct link to video on YouTube:


Visual Analytics, Tax Network, Tax Evasion Detection, Anomaly detection, Multidimensional data


Tax evasion is a serious economic problem for many countries, as it can undermine the government's tax system and lead to an unfair business competition environment. Recent research has applied data analytics techniques to analyze and detect tax evasion behaviors of individual taxpayers. However, they failed to support the analysis and exploration of the uprising related party transaction tax evasion (RPTTE) behaviors (e.g., transfer pricing), where a group of taxpayers is involved. In this paper, we present TaxThemis, an interactive visual analytics system to help tax officers mine and explore suspicious tax evasion groups through analyzing heterogeneous tax-related data. A taxpayer network is constructed and fused with the trade network to detect suspicious RPTTE groups. Rich visualizations are designed to facilitate the exploration and investigation of suspicious transactions between related taxpayers with profit and topological data analysis. Specifically, we propose a calendar heatmap with a carefully-designed encoding scheme to intuitively show the evidence of transferring revenue through related party transactions. We demonstrate the usefulness and effectiveness of TaxThemis through two case studies on real-world tax-related data, and interviews with domain experts.