FlowScope: Spotting Money Laundering Based on Graphs

• DOI: 10.1609/aaai.v34i04.5906
• Publication Date: 2020-06-29T21:36:48Z
• Authors: Xiangfeng Li; Shenghua Liu; Zifeng Li; Xiaotian Han; Chuan Shi; Bryan Hooi; He Huang; Xueqi Cheng
• Keywords: Money Laundering; Spotting

Given a graph of the money transfers between accounts bank, how can we detect laundering? Money laundering refers to criminals using bank's services move massive amounts illegal untraceable destination accounts, in order inject their into legitimate financial system. Existing fraud detection approaches focus on dense subgraph detection, without considering fact that involves high-volume flows funds through chains bank thereby decreasing accuracy. Instead, propose model transactions multipartite graph, and complete flow from source scalable algorithm, FlowScope. Theoretical analysis shows FlowScope provides guarantees terms amount fraudsters transfer being detected. outperforms state-of-the-art baselines accurately detecting involved laundering, both injected real-world data settings.