基于图卷积神经网络的洗钱行为识别研究

Research on Recognition of Money Laundering Behavior Based on Graph Convolutional Neural Network

洗钱犯罪影响到金融业健康发展,严重破坏经济秩序。银行体系是洗钱犯罪的高发领域,但大多数分行的反洗钱工作主要依据既有案例手动筛查,耗时耗力。大数据和AI技术的快速发展,使建立大数据和人工智能平台成为可能,完善补充现有反洗钱规则,缩小审查范围,降低审查成本,全面赋能银行业务。本次研究致力于探索在构造关联网络的基础上应用图卷积神经网络,建立反洗钱模型。

Money laundering crimes affect the healthy development of the financial industiy and severely disrupt the economic order.The banking system is the main channel for money laundering crimes,but the AML of most branches relies on manually screening for the suspected on the basis of existing cases,which is time-consuming and labor-intensive.The rapid development of big data and AI technology has made it possible to establish a platfbnn,which complements existing AML rules,narrows the scope of review;reduces review costs,and promotes related technologies to fully empower banking business.This research is devoted to exploring the application of graph convolutional neural network on the basis of constructing an association network to establish an AML model.