计算机科学

计算机科学 ›› 2025, Vol. 52 ›› Issue (10): 60-69.doi: 10.11896/jsjkx.250800009

• 数智赋能金融科技前沿 • 上一篇    下一篇

基于时序图神经网络的资产管理反洗钱检测方法

徐鑫1, 朱鸿斌2, 谌杰2, 李青汶3, 张霄蓉1, 吕智慧2   

  1. 1 上海立信会计金融学院计算机与人工智能学院 上海 201209
    2 复旦大学计算与智能创新学院 上海 200433
    3 上海财经大学信息管理与工程学院 上海 200433
  • 收稿日期:2025-08-04 修回日期:2025-09-13 出版日期:2025-10-15 发布日期:2025-10-14
  • 通讯作者: 张霄蓉(zhangxiaorongzj@163.com)
  • 作者简介:(20180152@lixin.edu.cn)
  • 基金资助:
    国家自然科学基金面上项目(61873309);国家重点研发计划(2023YFC3305304);上海科技创新行动计划(24692111100)

Anti-money Laundering Detection Method for Asset Management Based on Temporal Graph Neural Networks

XU Xin1, ZHU Hongbin2, CHEN Jie2, LI Qingwen3, ZHANG Xiaorong1, LYU Zhihui2   

  1. 1 School of Computing and Artificial Intelligence,Shanghai Lixin University of Accounting and Finance,Shanghai 201209,China
    2 College of Computer Science and Artificial Intelligence,Fudan University,Shanghai 200433,China
    3 School of Information Management and Engineering,Shanghai University of Finance and Economics,Shanghai 200433,China
  • Received:2025-08-04 Revised:2025-09-13 Online:2025-10-15 Published:2025-10-14
  • About author:XU Xin,born in 1983,Ph.D,associate professor,is a member of CCF(No.J2031M).His main research interest is financial data analysis.
    ZHANG Xiaorong,born in 1994,Ph.D,assistant professor.Her main research interest is data mining.
  • Supported by:
    National Natural Science Foundation of China(61873309),National Key Research and Development Program of China(2023YFC3305304) and Shanghai Science and Technology Innovation Action Plan(24692111100).

PDF (PC)

4

摘要： 资产管理行业因高频且灵活的资金操作方式,已成为洗钱活动的重要目标。然而,资产管理行业中交易结构的稀疏性、客户间隐性资金流转路径的复杂性,以及交易行为的非统一特征,使得传统显式关系的图建模方法难以有效应对这些挑战。针对上述问题,提出了一种基于时序图神经网络的资产管理反洗钱检测框架(AM-GAML)。该框架通过融合时序模型与图神经网络,构建时间-结构联合嵌入表示,并设计了基于隐式交互关系的图生成机制,能够充分挖掘交易记录中的弱关联特征并捕捉客户间复杂的交易行为模式。在真实交易数据集上的实验验证了AM-GAML在准确率、召回率、F1-score和AUPRC等多个关键指标上显著优于多个先进方法,尤其在少数类识别和泛化能力方面表现突出。所提方法为资产管理行业的反洗钱检测提供了高效且可靠的解决方案,并为复杂金融场景下的风险防控研究提供了有力支持。

关键词: 资产管理, 反洗钱, 时序模型, 图神经网络, 交易行为分析

Abstract: The asset management industry,characterized by high-frequency and flexible financial operations,has become a primary target for money laundering activities.However,the sparsity of transaction structures,the complexity of implicit fund transfer paths between accounts,and the non-uniform characteristics of transaction behaviors pose significant challenges for traditional graph modeling methods based on explicit relationships.To address these issues,this paper proposes an Anti-money Laundering Detection Framework for Asset Management based on Temporal Graph Neural Networks(AM-GAML).By integrating temporal models with graph neural networks,the proposed framework constructs a joint temporal-structural embedding representation and designs a graph generation mechanism based on implicit interaction relationships.This enables the framework to effectively capture weakly correlated features in transaction records and uncover complex transaction behavior patterns between users.Experimental results on a real-world transaction dataset demonstrate that AM-GAML significantly outperforms several advanced approaches in terms of accuracy,recall,F1-score,and AUPRC.The framework excels particularly in minority class detection and generalization ability.The proposed method provides an efficient and reliable solution for anti-money laundering detection in the asset management industry and offers valuable support for risk prevention and control research in complex financial scenarios.

Key words: Asset management,Anti-money laundering,Temporal model,Graph neural network,Transaction behavior analysis

中图分类号: 

  • TP181
[1]CHEN Z,VAN KHOA L D,TEOH E N,et al.Machine learning techniques for anti-money laundering(AML) solutions in suspicious transaction detection:A review[J].Knowledge and Information Systems,2018,57:245-285.
[2]NICHOLLS J,KUPPA A,LE-KHAC N A.Financial cyber-crime:A comprehensive survey of deep learning approaches to tackle the evolving financial crime landscape[J].IEEE Access,2021,9:163965-163986.
[3]ZHANG Y,TRUBEY P.Machine learning and samplingscheme:An empirical study of money laundering detection[J].Computational Economics,2019,54(3):1043-1063.
[4]JULLUM M,LØLAND A,HUSEBY R B,et al.Detecting money laundering transactions with machine learning[J].Journal of Money Laundering Control,2020,23(1):173-186.
[5]QIN Z P,ZHOU Y T,LI Z.Bank transaction fraud detection method based on graph neural network[J].Computer Science,2024,51(S2):921-928.
[6]LIU X,WANG X G.Probabilistic graphical model based ap-proach for bank telecommunication fraud detection[J].Compu-ter Science,2018,45(7):122-128,134.
[7]CHENG D,YE Y,XIANG S,et al.Anti-money laundering by group-aware deep graph learning[J].IEEE Transactions on Knowledge and Data Engineering,2023,35(12):12444-12457.
[8]LIANG Y,WU W,LIANG R,et al.A plug-and-play data-driven approach for anti-money laundering in bitcoin[J].Expert Systems with Applications,2025,266:126072.
[9]LORENZ J,SILVA M I,APARÍCIO D,et al.Machine learning methods to detect money laundering in the bitcoin blockchain in the presence of label scarcity[C]//Proceedings of the First ACM International Conference on AI in Finance.2020:1-8.
[10]LIU L,LI X,LAN T,et al.A survey on anti-money laundering techniques in blockchain systems[J].Strategic Study of Chinese Academy of Engineering,2025,27(2):287-303.
[11]LIU K Y,YU T T.An improved support-vector network model for anti-money laundering[C]//2011 Fifth International Conference on Management of E-Commerce and E-Go-vernment.IEEE,2011:193-196.
[12]BAKRY A N,ALSHARKAWY A S,FARAG M S,et al.Combating financial crimes with unsupervised learning techniques:Clustering and dimensionality reduction for anti-Money Laundering[J].arXiv:2403.00777,2024.
[13]TATULLI M P,PALADINI T,D'ONGHIA M,et al.HAM-LET:A transformer based approach for money laundering detection[C]//International Symposium on Cyber Security,Cryptology,and Machine Learning.Springer,2023:234-250.
[14]JENSEN R I T,IOSIFIDIS A.Qualifying and raising anti-money laundering alarms with deep learning[J].Expert Systems with Applications,2023,214:119037.
[15]CARDOSO M,SALEIRO P,BIZARRO P.Laundrograph:Self-supervised graph representation learning for anti-money laundering[C]//Proceedings of the Third ACM International Confe-rence on AI in Finance.2022:130-138.
[16]EDDIN A N,BONO J,APARÍCIO D,el al.Anti-money laundering alert optimization using machine learning with graphs[J].arXiv:2112.07508,2021.
[17]LI X,CHEN L.Graph anomaly detection with domain-Agnostic pre-training and few-shot adaptation[C]//2024 IEEE 40th International Conference on Data Engineering(ICDE).IEEE,2024:2667-2680.
[18]WEBER M,DOMENICONI G,CHEN J,et al.Anti-money laundering in bitcoin:Experimenting with graph convolutional networks for financial forensics[J].arXiv:1908.02591,2019.
[19]SCHOTT P A.Reference guide to anti-money laundering and combating the financing of terrorism[M].Washington,DC:World Bank Publications,2006.
[20]FORCE F A T.Risk-based approach guidance for the banking sector[R/OL].Paris:FATF,2014:48.https://www.fatf-gafi.org/.
[21]VAN DUYNE P C,SOUDIJN M R.Crime-money in the financial system-What we fear and what we know[M]//Transna-tional Criminology Manual:Volume 2.Nijmegen:Wolf Legal Publishers,2010.
[22]HEIDARINIA N,HAROUNABADI A,SADEGHZADEH M.An intelligent anti-money laundering method for detecting risky users in the banking systems[J].International Journal of Computer Applications,2014,97(22):35-39.
[23]EL-BANNA M M,KHAFAGY M H,EL KADI H M.Smurfdetector:a detection technique of criminal entities involved in money laundering[C]//2020 International Conference on Innovative Trends in Communication and Computer Engineering(ITCE).2020:64-71.
[24]CAI T,LUO S,XU K,et al.Graphnorm:A principled approach to accelerating graph neural network training[C]//International Conference on Machine Learning.2021:1204-1215.
[25]BRODY S,ALON U,YAHAV E.How attentive are graph attention networks?[J].arXiv:2105.14491,2021.
[26]HAMILTON W,YING Z,LESKOVEC J.Inductive representation learning on large graphs[C]//Proceedings of the 31st International Confernce on Neural Information Processing Systems.2017:1025-1035.
[27]HE K,ZHANG X,REN S,et al.Deep residual learning for image recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2016:770-778.
[28]XU K,LI C,TIAN Y,et al.Representation learning on graphs with jumping knowledge networks[C]//International Confe-rence on Machine Learning.2018:5453-5462.
[29]CORTES C,VAPNIK V.Support-vector networks[J].Machine Learning,1995,20:273-297.
[30]FRIEDMAN J H.Greedy function approximation:a gradientboosting machine[J].Annals of Statistics,2001,29(5):1189-1232.
[31]CHEN T,GUESTRIN C.Xgboost:A scalable tree boosting system[C]//Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.2016:785-794.
[32]HOCHREITER S,SCHMIDHUBER J.Long short-term memory[J].Neural Computation,1997,9(8):1735-1780.
[33]VELICKOVIC P,CUCURULL G,CASANOVA A,et al.Graph attention networks[J].Stat,2017,1050(20):10-48550.
[34]HU Z,DONG Y,WANG K,et al.Heterogeneous graph transformer[C]//Proceedings of the Web Conference.2020:2704-2710.
[35]SCHLICHTKRULL M,KIPF T N,BLOEM P,et al.Modeling relational data with graph convolutional networks[C]//European Semantic Web Conference.2018:593-607.
[36]FU X,ZHANG J,MENG Z,et al.Magnn:Metapath aggregated graph neural network for heterogeneous graph embedding[C]//Proceedings of the Web Conference.2020:2331-2341.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发