Method and device for detecting money laundering behavior in streaming graph, and medium

Abstract

The invention provides a money laundering behavior detection method and apparatus in a streaming graph, and a medium. The method comprises the steps of obtaining fund transaction information in a money laundering scene as graph data, and representing the graph data through a directed graph; when each fund flow arrives, judging whether the occurrence source account reaches a fund balance state after fund transfer-out, and whether the target account reaches the fund balance state after fund transfer-in; and respectively counting the number of times that the occurrence source account and the target account reach the fund balance state at the current moment and the number of times of effective fund transfer accumulated from the initial moment to the last fund balance state at the current moment, and inputting the number of times of effective fund transfer into an abnormal account evaluation model as statistical characteristics to carry out abnormal evaluation. According to the method, money laundering behaviors in different modes can be rapidly and accurately detected, and a suspicious money laundering account can be found.

Description

technical field [0001] The present invention relates to the technical field of abnormality detection, and in particular, to a method, device and medium for detecting money laundering behavior in a streaming graph suitable for streaming scenarios. Background technique [0002] Money laundering refers to the act of concealing the source of funds through improper activities and is a rapid process of frequent, chain transactions. Usually, in the process of money laundering, fraudsters will use multiple intermediate accounts to pool funds, mix them with legitimate transactions, and make the amount of the mixed transactions just below the detection threshold, so the detection of money laundering behavior becomes very difficult. In real-time transactions, how to design an effective detection method that can quickly and accurately detect such suspicious accounts and behavior patterns is the key point to be considered in money laundering detection. [0003] In order to introduce mor...

AI Technical Summary

Problems solved by technology

However, these streaming methods do not consider the characteristics of money laundering accounts, so they are not suitable for the detection of money laundering accounts

The FlowScope algorithm and the AutoAudit algorithm are proposed to detect money laundering flows in multipartite graphs, but this type of method cannot handle flow edges and cannot capture various money laundering behaviors well

[0005] Therefore, most of the existing money laundering detection algorithms are designed for static transaction data and cannot be applied to streaming scenarios

The anomaly detection algorithm based on the dynamic graph does not consider the characteristics of money laundering, so the detection accuracy is low

Images