计算机科学

计算机科学 ›› 2024, Vol. 51 ›› Issue (11A): 240200024-8.doi: 10.11896/jsjkx.240200024

• 信息安全 • 上一篇    下一篇

基于图神经网络的银行交易欺诈检测方法

秦忠飘1, 周亚同1, 李哲2   

  1. 1 河北工业大学电子信息工程学院 天津 300401
    2 河北工业大学数字经济产业研究院 石家庄050000
  • 出版日期:2024-11-16 发布日期:2024-11-13
  • 通讯作者: 周亚同(zyt@hebut.edu.cn)
  • 作者简介:(2678282628@qq.com)
  • 基金资助:
    京津冀基础研究合作专项(J210008,21JCZXJC00170,H2021202008);内蒙古自治区纪检监察大数据实验室开放课题(IMDBD202105)

Bank Transaction Fraud Detection Method Based on Graph Neural Network

QIN Zhongpiao1, ZHOU Yatong1, LI Zhe2   

  1. 1 School of Electronics and Information Engineering,Hebei University of Technology,Tianjin 300401,China
    2 Institute of Digital Economy Industry Research,Hebei University of Technology,Shijiazhuang 050000,China
  • Online:2024-11-16 Published:2024-11-13
  • About author:QIN Zhongpiao,born in 1999,postgra-duate,is a member of CCF(No.P6799G).His main research interest is fraud detection.
    ZHOU Yatong,born in 1973,Ph.D,professor.His main research interests include pattern recognition and machine learning.
  • Supported by:
    Special Foundation for Beijing Tianjin Hebei Basic Research Cooperation(J210008,21JCZXJC00170,H2021202008) and Inner Mongolia Discipline Inspection and Supervision Big Data Laboratory(IMDBD202105).

PDF (PC)

172

摘要： 随着电子支付的迅速发展,欺诈问题日益增多。传统欺诈检测方法受限于规则和特征工程,难以捕获复杂的交易模式。相反,基于图的方法虽然强调了数据的关系性,但通常忽略了特征工程的重要性。为了解决这个问题,提出一种端到端的电信诈骗检测方法——基于图神经网络的银行交易欺诈检测方法。该方法设计了一个针对图数据的特征工程,并利用融合模型对其进行训练。具体来说,使用过采样和设置节点权重的方式对银行交易数据进行不平衡处理,然后采用改进的自适应相似度边和节点度权重融合策略,构建用户交易图数据并挖掘交易节点间的潜在关联信息,最后综合局部特征和全局特征通过模型融合来弥补单一分类器的不足。实验结果表明,在广西玉林银行的交易数据中,所提模型对于交易欺诈数据的检测在F1分数、召回率、AUC 3个指标上相比GraphSAGE分别提升1.65,1.36,4.2个百分点,图数据构建时间缩短了80%左右,与其他主流的检测算法相比也取得了更高的检测精度。

关键词: 银行交易欺诈检测, 特征工程, 不平衡处理, 相似度边, 图神经网络

Abstract: With the rapid development of electronic payments,the fraud problem is increasing.Limited by rule and feature engineering,traditional fraud detection methods are difficult to capture complex transaction patterns.Conversely,graph-based me-thods often downplay the significance of feature engineering while highlighting the relational aspect of the data.In addition,few studies have examined the application of graph methods in the field of fraud detection for specific bank transaction data.To address this problem,this paper proposes an end-to-end telecom fraud detection method,a fraud detection method for banking tran-sactions based on graph neural networks.The proposed method designs a feature engineering for graph models and trains it using a fusion model.Specifically,oversampling and node weighting are used to address the unbalanced dataset.Next,a user transaction graph model is built utilizing an adaptive similarity edge and node degree weight fusion technique to construct a user transaction graph model and mine potential correlation information between transaction nodes.Furthermore,model fusion is employed to merge local and global variables to overcome the constraints of separate classifiers.Experimental results show that in Guangxi Yulin Bank transaction data,the proposed model for the detection of transaction fraud dataon the three indicators of F1 score,recall rate,and AUC is improved by 1.65%,1.36% and 4.2% compared to GraphSAGE,respectively.The model also achieves a reduction of approximately 80% in training time.In comparison to other mainstream detection algorithms,it exhibits higher detection accuracy.

Key words: Bank transaction fraud detection, Feature engineering, Unbalanced treatment, Similarity edge, Graph neural network

中图分类号: 

  • TP181
[1]BECKER R A,VOLINSKY C,WILKS A R.Fraud Detection inTelecommunications:History and Lessons Learned [J].Technometrics,2010,52(1):20-33.
[2]BARSON P,FIELD S,DAVEY N,et al.The detection of fraud in mobile phone networks [J].Neural Network World,1996,6(4):477-484.
[3]HILAS C S.Designing an expert system for fraud detection inprivate telecommunications networks [J].Expert Systems with Applications,2009,36(9):11559-69.
[4]ELMI A H,IBRAHIM S,SALLEHUDDIN R.Detecting SIM Box Fraud Using Neural Network[C]//Proceedings of the IT Convergence and Security.2012.
[5]ZHENG Y J,ZHOU X H,SHENG W G,et al.Generative adversarial network based telecom fraud detection at the receiving bank [J].Neural Networks,2018,102:78-86.
[6]YANG H B,MA Y C,LI J L.Telecommunication Fraud Identification System Based on Convolutional neural network [J].Telecommunications Technology,2019,(6):60-63,68.
[7]YANG H T,WANG H P,CHU X T,et al.Fake speech detection based on deep Convolutional neural network[J].Police Technology,2022,190(1):3336.
[8]LIU S,JI X,LIU C,et al.Extended resource allocation index for link prediction of complex network [J].Physica A:Statistical Mechanics and its Applications,2017,479:174-183.
[9]ZHENG Q Z,XU P F.Structure Learning of Gaussian Graphical Models with Latent Variables Based on Adaptive Penalties [J].Journal of Jilin University(Scinence Edition),2023,61(5):1056-1062.
[10]WEST D B.Introduction to graph theory [M].Prentice hall Upper Saddle River,2001.
[11]ZHANG J J,TANG Y C,JI S Y,et al,A telecom fraud identification method based on graph neural network [J].Electronics Science Technology and Application,2021,47(6):25-29,34.
[12]LIU M,LIAO J,WANG J,et al.AGRM:attention-based graph representation model for telecom fraud detection[C]//2019 IEEE International Conference on Communications(ICC 2019).IEEE.2019.
[13]XU B,SHEN H,SUN B,et al.Towards consumer loan fraud detection:Graph neural networks with role-constrained conditional random field[C]//Proceedings of the AAAI Conference on Artificial Intelligence.2021.
[14]LIANG T,ZENG G,ZHONG Q,et al.Credit risk and limits forecasting in e-commerce consumer lending service via multi-view-aware mixture-of-experts nets[C]//Proceedings of the 14th ACM international Conference on Web Search and Data Mining.2021.
[15]ZHANG Y,FAN Y,YE Y,et al.Key player identification in underground forums over attributed heterogeneous information network embedding framework[C]//Proceedings of the 28th ACM International Conference on Information and Knowledge Management.2019.
[16]DING K,ZHOU Q,TONG H,et al.Few-shot network anomaly detection via cross-network meta-learning[C]//Proceedings of the Web Conference.2021.
[17]DING K,SHU K,SHAN X,et al.Cross-domain graph anomaly detection [J].IEEE Transactions on Neural Networks and Learning Systems,2021,33(6):2406-2415.
[18]HAMILTON W,YING Z,LESKOVEC J.Inductive representation learning on large graphs [J].Advances in Neural Information Processing Systems,2017,30.
[19]VELIČKOVIĆ P,CUCURULL G,CASANOVA A,et al.Graph attention networks [J].arXiv:171010903,2017.
[20]WU Z,PAN S,CHEN F,et al.A comprehensive survey ongraph neural networks [J].IEEE Transactions on Neural Networks and Learning Systems,2020,32(1):4-24.
[21]HAMILTON W L,YING R,LESKOVEC J.Representationlearning on graphs:Methods and applications [J].arXiv:170905584,2017.
[22]ZHOU J,CUI G,HU S,et al.Graph neural networks:Areview of methods and applications [J].AI Open,2020,1:57-81.
[23]FISCHER A,BOTERO J F,BECK M T,et al.Virtual network embedding:A survey [J].IEEE Communications Surveys & Tutorials,2013,15(4):1888-1906.
[24]WU Z,PAN S,CHEN F,et al.A Comprehensive Survey onGraph Neural Networks [J].IEEE Transactions on Neural Networks and Learning Systems,2021,32(1):4-24.
[25]KANG J M.Distance Metrics [M]//Encyclopedia of GIS.Cham;Springer International Publishing.2017:483-484.
[26]CHOWDHURY G G.Introduction to modern information re-trieval [M].Facet Publishing,2010.
[27]CORTES C,VAPNIK V.Support-vector networks [J].Machine Learning,1995,20:273-97.
[28]KIPF T N,WELLING M.Semi-supervised classification withgraph convolutional networks [J].arXiv:1609.02907,2016.
[29]XU K,HU W,LESKOVEC J,et al.How powerful are graphneural networks? [J].arXiv:1810.00826,2018.
[30]TANG J,LI J,GAO Z,et al.Rethinking graph neural networks for anomaly detection[C]//Proceedings of the International Conference on Machine Learning.PMLR,2022.
[31]LIU Y,AO X,QIN Z,et al.Pick and choose:a GNN-based imbalanced learning approach for fraud detection[C]//Proceedings of the Web Conference.2021.
[32]HU X,CHEN H,CHEN H,et al.Mining Mobile NetworkFraudsters with Augmented Graph Neural Networks [J].Entropy,2023,25(1):150.
[33]DOU Y,LIU Z,SUN L,et al.Enhancing graph neural network-based fraud detectors against camouflaged fraudsters[C]//Proceedings of the 29th ACM International Conference on Information & Knowledge Management.2020.
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