计算机科学

计算机科学 ›› 2024, Vol. 51 ›› Issue (6A): 230600040-7.doi: 10.11896/jsjkx.230600040

• 信息安全 • 上一篇    下一篇

融合多源图特征的Kcore-GCN反欺诈算法研究

刘炜1, 宋友1, 卓佩妍1, 仵伟强2, 廉鑫2   

  1. 1 北京航空航天大学软件学院 北京 100191
    2 渤海银行 天津 300070
  • 发布日期:2024-06-06
  • 通讯作者: 宋友(songyou@buaa.edu.cn)
  • 作者简介:(buaaliuwei@buaa.edu.cn)
  • 基金资助:
    河北省重点研发计划(21310101D)

Study on Kcore-GCN Anti-fraud Algorithm Fusing Multi-source Graph Features

LIU Wei1, SONG You1, ZHUO Peiyan1, WU Weiqiang2, LIAN Xin2   

  1. 1 College of Software,Beihang University,Beijing 100191,China
    2 China Bohai Bank Co.,Ltd,Tianjin 300070,China
  • Published:2024-06-06
  • About author:LIU Wei,born in 1998,master.His main research interests include data mining and graph neural networks.
    SONG You,born in 1973,Ph.D,professor.His main reaserch interests include software engineering,big data analysis,technology finance,and so on.
  • Supported by:
    National Key Research and Development Program of Hebei Province,China(21310101D).

PDF (PC)

245

摘要： 金融欺诈行为给社会带来了许多负面影响,针对金融欺诈行为,多种人工智能与金融反欺诈算法被提出并应用于实际反欺诈业务场景,取得了不错的成绩。这些反欺诈算法或从用户个体的角度进行欺诈检测,或从节点与网络的拓扑关系的角度进行欺诈检测,或通过学习节点的图嵌入式表示进行欺诈检测,出发角度较为局限,无法进行完备的欺诈检测分析。针对上述问题,设计了一种基于融合多源图特征的Kcore图卷积神经网络反欺诈算法,该算法的创新性在于能够高效挖掘网络中节点层级的拓扑关系与全局网络层次的拓扑关系来构建宽领域的特征体系,并通过基于Kcore算法的图卷积神经网络完成深层次图结构特征的传播与聚合,最终完成欺诈风险的检测。实验效果表明,该方法相较于相关机器学习算法与图神经网络算法在相关评价指标上均有较大的提升,其中较LightGBM算法有12%的AUC值提升,较GCN算法有6%的AUC值提升。

关键词: 机器学习, 图表示学习, 图神经网络, 金融欺诈检测

Abstract: Financial fraud has brought many negative impacts to society,and a variety of AI and financial anti-fraud algorithms have been applied to practical anti-fraud business scenarios and have achieved good results.These anti-fraud algorithms either perform fraud detection from the perspective of individual users,or perform fraud detection from the perspective of topological relationship between nodes and network,or perform fraud detection by learning the graph embedded representation of nodes,which are limited in their starting perspectives and cannot perform a complete fraud detection analysis.To address the above problems,this paper designs a Kcore graph convolutional neural network anti-fraud algorithm based on the fusion of multi-source graph features.The innovation of this algorithm lies in the fact that it can efficiently mine the topological relationships at the node level in the network and the topological relationships at the global network level to build a wide-field feature system,and complete the propagation and aggregation of deep-level graph structure features through the graph convolutional neural network based on the Kcore algorithm The final result is the detection of fraud risk.Experimental results show that the method has a large improvement in the evaluation indexes compared with related machine learning algorithms and graph neural network algorithms,including a 12% improvement in the AUC value compared with LightGBM algorithm and a 6% improvement in the AUC value compared with GCN algorithm.

Key words: Machine learning, Graph representation learning, Graph neural network, Financial fraud detection

中图分类号: 

  • TP183
[1]中国工商银行金融科技研究院安全攻防实验室,中国信息通信研究院云计算与大数据研究所.数字金融反欺诈技术应用分析报告[EB/OL].https://www.freebuf.com/articles/paper/325372.html.
[2]GU X D.Research and application of large-scale rule reasoning engine based on rule algorithm[D].Nanjing University,2013.
[3]FORGY C L.Rete:A fast algorithm for the many pattern/many object pattern match problem[J].Artificial Intelligence,1982,19(1):17-37.
[4]LIU D,GU T,XUE J P.Rule Engine Based onimprovement Rete algorithm[C]//The 2010 International Conference on Apperceiving Computing and Intelligence Analysis Proceeding.Chengdu,2010:346-349.
[5]SUN Y,WU T Y,ZHAO G,et al.Efficient Rule Engine forSmart Building Systems[C]//IEEE Transactions on Compu-ters.2015:1658-1669.
[6]CHATTOPADHYAY S,BANERJEE A,BANERJEE N.AScalable Rule Engine Architecture for Service Execution Frameworks[C]//2016 IEEE International Conference on Services Computing(SCC).San Francisco,CA,2016:689-696.
[7]TU E,LU J.Evaluating credit risk and loan performance in online Peer-to-Peer(P2P) lending[J].Applied Economics,2015,47(1):54-70.
[8]KRUPPA J,SCHWARZ J,ARMINGER G,et al.ConsumerCredit Risk:Individual Probability Estimates Using Machine Learning[J].Expert Systems with Applications,2013,40(13):5125-5131.
[9]YU J X,QIAN W N,LU H J,et al.Finding centric local outliers in categorical/numerical spaces[J].Knowledge and Information Systems,2006,9(3):309-338.
[10]ROBINSON W N,ARIA A.Sequential fraud detection for prepaid cards using hidden Markov model divergence[J].Expert Systems with Applications,2018,91:235-251.
[11]LUCAS Y,PORTIER P E,LAPORTE L,et al.Towards automatedfeature engineering for credit card fraud detection using multi-perspective HMMs[J].Future Generation Computer Systems-the International Journal of Escience,2020,102:393-402.
[12]WANG J C.Rredit Risk Analysis of Bank Users Based on Machine Learning Algorithm[D].Tianjin:Nankai University,2021.
[13]DREZ·EWSKI R,SEPIELAK J,FILIPKOWSKI W.The application of social network analysis algorithms in a system suppor-ting money laundering detection[J].Information Sciences,2015,295:18-32.
[14]AKOGLU L,MCGLOHON M,FALOUTSOS C.Oddball:Spotting anomalies in weighted graphs[C]//Pacific-Asia Conference on Knowledge Discovery and Data Mining.Berlin,Heidelberg:Springer,2010:410-421.
[15]WANG K,CHEN D.Graph structure based anomaly behavior detection[C]//2nd International Conference on Computer Engineering,Information Science & Application Technology(ICCIA 2017).Atlantis Press,2016:531-538.
[16]KIPF T N,WELLING M.Semi-supervised classification withgraph convolutional networks[J].arXiv:1609.02907,2016.
[17]XU X,YURUK N,FENG Z,et al.Scan:a structural clustering algorithm for networks[C]//Proceedings of the 13th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.2007:824-833.
[18]SHI M,YANG Y F,ZHU X Q,et al.MultiClass Imbalanced Graph Convolutional Network Learning[C]//IJCAI.2020.
[19]LIU J,XU C,YIN C,et al.K-core based temporal graph convolutional network for dynamic graphs[J].IEEE Transactions on Knowledge and Data Engineering,2020,34(8):3841-3853.
[20]PEROZZI B,AL-RFOU R,SKIENA S.Deepwalk:Online lear-ning of social representations[C]//Proceedings of the 20th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.2014:701-710.
[21]GROVER A,LESKOVEC J.node2vec:Scalable feature learning for networks[C]//Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mi-ning.2016:855-864.
[22]LI Y,LIU X,WANG C.Research on Link Prediction under the Structural Features of Attention Stream Network[C]//2021 IEEE Asia-Pacific Conference on Image Processing,Electronics and Computers(IPEC).IEEE,2021:148-154.
[23]KONG Y X,SHI G Y,WU R J,et al.k-core:Theories and applications[J].Physics Report,2019,832:1-32.
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