Representation Learning on Graphs to Identifying Circular Trading in Goods and Services Tax
This addresses tax evasion detection for government authorities, but it appears incremental as it applies existing graph techniques to a specific domain problem.
The paper tackles the problem of identifying circular trading, a form of tax evasion in Goods and Services Tax, by using graph representation learning to detect communities of fraudulent traders and illegitimate transactions, with testing on real-life data from Telangana, India uncovering several such communities.
Circular trading is a form of tax evasion in Goods and Services Tax where a group of fraudulent taxpayers (traders) aims to mask illegal transactions by superimposing several fictitious transactions (where no value is added to the goods or service) among themselves in a short period. Due to the vast database of taxpayers, it is infeasible for authorities to manually identify groups of circular traders and the illegitimate transactions they are involved in. This work uses big data analytics and graph representation learning techniques to propose a framework to identify communities of circular traders and isolate the illegitimate transactions in the respective communities. Our approach is tested on real-life data provided by the Department of Commercial Taxes, Government of Telangana, India, where we uncovered several communities of circular traders.