计算机科学 ›› 2026, Vol. 53 ›› Issue (9): 439-450.doi: 10.11896/jsjkx.260100159
• 信息安全 • 上一篇
江航宇, 曹怀虎, 周凯, 陈福, 苏瑞
JIANG Hangyu, CAO Huaihu, ZHOU Kai, CHEN Fu, SU Rui
摘要： 近年来,随着互联网服务的发展,欺诈行为逐渐增多且手法不断演变。图神经网络因其强大的结构信息建模能力而被广泛应用于欺诈检测任务。然而,现有大多数方法基于“同质性假设”,这在实际欺诈检测应用场景中常常不成立,因为欺诈行为本质上是异质性的——欺诈者通过连接正常节点来伪装自身。同时,节点类别不平衡问题显著存在,但边关系的不平衡及其潜在信息的利用仍未得到充分关注。为此,提出一种面向异质性与不平衡图的欺诈检测方法。具体而言,设计可学习的双通道图卷积滤波器,自适应地聚合邻居节点的低频和高频信号,实现对同质性和异质性边上节点特征的有效建模;同时,引入类别感知的节点与边采样器,以缓解图的节点和边不平衡的问题,并将采样的边作为辅助监督信号指导滤波器训练。在4个真实数据集上的实验结果表明,所提方法在提升欺诈节点识别能力和模型鲁棒性方面均优于各类典型的基线方法。
中图分类号:
| [1] MOTIE S,RAAHEMI B.Financial Fraud Detection usingGraph Neural Networks:A Systematic Review[J].Expert Systems with Applications,2024,240:122156. [2] LIU B,SUN X G,MENG Q,et al.Nowhere to hide:Online Rumor Detection Based on Retweeting Graph Neural Networks[J].IEEE Transactions on Neural Networks and Learning Systems,2024,35(4):4887-4898. [3] YANG J,ZHANG R,CHENG Z,et al.Grad:Guided RelationDiffusion Generation for Graph Augmentation in Graph Fraud Detection[C] //Proceedings of the ACM on Web Conference.2025:5308-5319. [4] QIAO H Z,TONG H,AN B,et al.Deep Graph Anomaly Detection:A Survey and New Perspectives[J].IEEE Transactions on Knowledge and Data Engineering,2025,37(9):5106-5126. [5] JU W,YI S Y,WANG Y F,et al.A Survey of Graph Neural Networks in Real World:Imbalance,Noise,Privacy and OOD Challenges[J/OL].IEEE Transactions on Pattern Analysis and Machine Intelligence,2025,https://doi.org/10.1109/TPAMI.2025.3630673. [6] YU P J,LI X,QI J P,et al.Multiplex Heterogenous Graph Neural Network for Node Classification[J].Journal of Software,2026,37(2):716-731. [7] PAN J,LIU Y,ZHENG X,et al.A Label-free Heterophily-guided Approach for Unsupervised Graph Fraud Detection[C] //Proceedings of the AAAI Conference on Artificial Intelligence.2025:12443-12451. [8] LIU Z,LI Y,CHEN N,et al.A Survey of Imbalanced Learning on Graphs:Problems,Techniques,and Future Directions[J].IEEE Transactions on Knowledge and Data Engineering,2025,37(6):3132-3152. [9] JIN W,MA Y,LIU X R,et al.Graph Structure Learning for Robust Graph Neural Networks[C] //Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Disco-very and Data Mining.2020:66-74. [10] LI S,KIM D,WANG Q.Restructuring Graph for Higher Homophily via Adaptive Spectral Clustering[C] //Proceedings of the AAAI Conference on Artificial Intelligence.2023:8622-8630. [11] ZHU J,YAN Y,ZHAO L,et al.Beyond Homophily in Graph Neural Networks:Current Limitations and Effective Designs[C] //Advances in Neural Information Processing Systems.2020:7793-7804. [12] SURESH S,BUDDE V,NEVILLE J,et al.Breaking the Limit of Graph Neural Networks by Improving the Assortativity of Graphs with Local Mixing Patterns[C] //Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining.2021:1541-1551. [13] LI X,ZHU R,CHENG Y,et al.Finding Global Homophily in Graph Neural Networks When Meeting Heterophily[C] //International Conference on Machine Learning.2022:13242-13256. [14] CHIEN E,PENG J,LI P,et al.Adaptive Universal Generalized PageRank Graph Neural Network[C] //Proceedings of the 9th International Conference on Learning Representations.2021. [15] BO D,WANG X,SHI C,et al.Beyond Low-frequency Information in Graph Convolutional Networks[C] //Proceedings of the AAAI Conference on Artificial Intelligence.2021:3950-3957. [16] SHI F,CAO Y,SHANG Y,et al.H2-FDetector:A GNN-based Fraud Detector with Homophilic and Heterophilic Connections[C] //Proceedings of the ACM Web Conference.2022:1486-1494. [17] DU L,SHI X Z,FU Q,et al.GBK-GNN:Gated Bi-Kernel Graph Neural Networks for Modeling Both Homophily and Heterophily[C] //Proceedings of the ACM Web Conference.2022:1550-1558. [18] HE M,WEI Z,XU H.BernNet:Learning Arbitrary Graph Spectral Filters via Bernstein Approximation[C] //Advances in Neural Information processing Systems.2021:14239-14251. [19] SHI M,TANG Y,ZHU X,et al.Multi-class Imbalanced Graph Convolutional Network Learning[C] //Proceedings of the Twenty-ninth International Joint Conference on Artificial Intelligence.2020. [20] JUAN X,ZHOU F,WANG W,et al.INS-GNN:ImprovingGraph Imbalance Learning with Self-supervision[J].Information Sciences,2023,637:118935. [21] LI X,FAN Z,HUANG F,et al.Graph Neural Network with Curriculum Learning for Imbalanced Node Classification[J].Neurocomputing,2024,574:127229. [22] HAJEK P,NOVOTNY J,MUNK M.Financial Statement Fraud Detection using Topic-driven Financial Sentiment Analysis[J].Decision Support Systems,2026,203:114615. [23] SISODIA D,SISODIA D S.A Transfer Learning FrameworkTowards Identifying Behavioral Changes of Fraudulent Publishers in Pay-per-click Model of Online Advertising for Click Fraud Detection[J].Expert Systems with Applications,2023,232:120922. [24] LI Z,LIU G,JIANG C.Deep Representation Learning with Full Center Loss for Credit Card Fraud Detection[J].IEEE Transactions on Computational Social Systems,2020,7(2):569-579. [25] FANAI H,ABBASIMEHR H.A Novel Combined Approachbased on Deep Autoencoder and Deep Classifiers for Credit Car Fraud Detection[J].Expert Systems with Applications,2023,217:119562. [26] XIE Y,LIU G J,YAN C,et al.Time-aware Attention-based Gated Network for Credit Card Fraud Detection by Extracting Transactional Behaviors[J].IEEE Transactions on Computational Social Systems,2023,10(3):1004-1016. [27] LIU Z Q,CHEN C,YANG X,et al.Heterogenous Graph Neural Networks for Malicious Account Detection[C] //Proceedings of the 27th ACM International Conference on Information and Knowledge Management.2018:2077-2085. [28] WANG J,WEN R,WU C,et al.FdGars:Frauster Detection via Graph Convolutional Networks in Online App Review System[C] //Companion Proceedings of the 2019 World Wide Web Conference.2019:310-316. [29] LIU Z W,DOU Y T,YU P S,et al.Alleviating the Inconsistency Problem of Applying Graph Neural Network to Fraud Detection[C] //Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval.2020:1569-1572. [30] DOU Y T,LIU Z W,SUN L,et al.Enhancing Graph NeuralNetwork-based Fraud Detectors against Camouflaged Fraudsters[C] //Proceedings of the 29th ACM International Conference on Information and Knowledge Management.2020:315-324. [31] LIU Y,AO X,QIN Z,et al.Pick and Choose:A GNN-based Imbalanced Learning Approach for Fraud Detection[C] //Procee-dings of the ACM Web Conference.2021:3168-3177. [32] ZHANG G,WU J,YANG J,et al.FRAUDRE:Fraud Detection Dual-Resistant to Graph Inconsistency and Imbalance[C] //2021 IEEE International Conference on Data Mining(ICDM).2021:867-876. [33] TANG J,LI J,GAO Z,et al.Rethinking Graph Neural Networks for Anomaly Detection[C] //International Conference on Machine Learning.2022:21076-21089. [34] WANG Y,ZHANG J,HUANG Z,et al.Label Information Enhanced Fraud Detection against Low Homophily in Graphs[C] //Proceedings of the ACM Web Conference.2023:406-416. [35] GAO Y,WANG X,HE X,et al.Addressing Heterophily inGraph Anomaly Detection:A Perspective of Graph Spectrum[C] //Proceedings of the ACM Web Conference.2023:1528-1538. [36] WU B,YAO X,ZHANG B,et al.SplitGNN:Spectral GraphNeural Network for Fraud Detection against Heterophily[C] //Proceedings of the 32nd ACM International Conference on Information and Knowledge Management.2023:2737-2746. [37] GAO Y,WANG X,HE X,et al.Alleviating Structural Distribution Shift in Graph Anomaly Detection[C] //Proceedings of the Sixteenth ACM International Conference on Web Search and Data Mining.2023:357-365. [38] ZHUO W,LIU Z,HOOI B,et al.Partitioning Message Passing for Graph Fraud Detection[C] //Proceedings of the Twelfth International Conference on Learning Representations.2024. [39] RAYANA S,AKOGLU L.Collective Opinion Spam Detection:Bridging Review Networks and Metadata[C] //Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.2015:985-994. [40] MCAULEY J J,LESKOVEC J.From Amateurs to Connois-seurs:Modeling the Evolution of User Expertise Through Online Reviews[C] //Proceedings of the 22nd International Confe-rence on World Wide Web.2013:897-908. [41] WEBER M,DOMENICONI G,CHEN J,et al.Anti-moneyLaundering in Bitcoin: Experimenting with Graph Convolutional Networks for Financial Forensics[J].arXiv:1908.02591,2019. [42] KIPF T N,WELLING M.Semi-supervised Classification withGraph Convolutional Networks[C] //Proceedings of the 5th International Conference on Learning Representations.2017. [43] LI P,YU H,LUO X.Context-Aware Graph Neural Network for Graph-based Fraud Detection with Extremely Limited Labels[C] //Proceedings of the AAAI Conference on Artificial Intelligence.2025:12112-12120. [44] WU J,LIU X,CHENG D,et al.Safeguarding Fraud Detection from Attacks:A Robust Graph Learning Approach[C] //IJCAI.2024:7500-7508. [45] ZHAO T,ZHANG X,WANG S.GraphSMOTE:ImbalancedNode Classification on Graphs with Graph Neural Networks[C] //Proceedings of the 14th ACM International Conference on Web Search and Data Mining.2021:833-841. [46] PARK J,SONG J,YANG E.GraphENS:Neighbor-Aware Ego Network Synthesis for Class-Imbalanced Node Classification[C] //International Conference on Learning Representations.2021. [47] LI K D,YANG T M,ZHOU M,et al.SEFraud:Graph-based Self-explainable Fraud Detection via Interpretative Mask Lear-ning[C] //Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining.2024:5329-5338. [48] DUAN M,HE D,ZHENG T,et al.Global Attribute-Association Pattern Aggregation for Graph Fraud Detection[C] //Proceedings of the AAAI Conference on Artificial Intelligence.2025:11616-11624. |
|
||
首页