Dynamic Heterogeneous Network Representation Learning for Fraud Detection in Auto Insurance

doi:10.3778/j.issn.1002-8331.2308-0078

Abstract

Abstract: Since the challenges inspired by the diverse and heterogeneous of the data and the large amount of historical data, a dynamic heterogeneous network representation learning method for fraud detection in the auto insurance is proposed. The graph is utilized to represent different structure and rich attribute nodes as vectors, and traditional machine learning algorithm is employed for fraud detection. Firstly, five random walk rules are designed based on the fraud types in the auto insurance, which provide multiple perspectives to describe fraud events. Secondly, a dynamic heterogeneous network node selection method is proposed to identify nodes relevant to newly collected auto insurance cases and the frequency of node in historical cases is calculated. The random walk paths and vector representations of these nodes are dynamically updated at a new timestamp. Finally, the effectiveness of the proposed algorithm is tested using the real auto insurance data, considering the fraud detection rate, fraud alarm rate, accuracy of fraud detection, running time, number of nodes and window size.

Key words: dynamic, heterogeneous network, representation learning, random walk, fraud detection

摘要： 针对车险理赔数据呈现多源异构形态及历史数据量较大的问题，提出一种基于动态异质网络表征学习的车险理赔反欺诈识别方法。利用图结构将车险理赔案件中不同结构多种属性的节点统一表示成向量形式，再利用传统的机器学习方法实现欺诈识别。根据车险理赔案件的欺诈类型，设计五条相应的随机游走规则，从多个角度描述车险欺诈事件；提出一种动态异质网络节点选择方法，辨识与新采集车险案件相关的节点，并计算历史案件中出现频次较高的节点，在新的时间戳动态更新这部分节点的随机游走路径及节点的向量表示。利用真实的车险理赔数据验证所提出算法的有效性，分别从欺诈的识别率、误报率、准确率、模型运行时间、节点选择数量以及模型窗口大小方面实现。

关键词: 动态性, 异质网络, 表征学习, 随机游走, 欺诈识别

PAN Yijun, LIANG Bian, ZHANG Long, NA Chongning. Dynamic Heterogeneous Network Representation Learning for Fraud Detection in Auto Insurance[J]. Computer Engineering and Applications, 2024, 60(24): 322-330.

潘怡君, 梁变, 张泷, 那崇宁. 动态异质网络表征学习的车险理赔反欺诈识别研究[J]. 计算机工程与应用, 2024, 60(24): 322-330.

References

[1] HILAL W, GADSDEN S, YAWNEY J. Financial fraud: a review of anomaly detection techniques and recent advances[J]. Expert Systems with Applications, 2022, 193: 116429.
[2] CAO L. AI in finance: challenges, techniques, and opportunities[J]. ACM Computing Surveys (CSUR), 2022, 55(3): 1-38.
[3] CHEN T, TSOURAKAKIS C. Antibenford subgraphs: unsupervised anomaly detection in financial networks[C]//Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 2022: 2762-2770.
[4] REN J. Research on financial investment decision based on artificial intelligence algorithm[J]. IEEE Sensors Journal, 2021, 21(22): 25190 - 25197.
[5] AHMED M, MAHMOOD A, ISLAM M. A survey of anomaly detection techniques in financial domain[J]. Future Generation Computer Systems, 2016, 55: 278-288.
[6] BURNS N, KEDIA S. The impact of performance-based compensation on misreporting[J]. Journal of Financial Economics, 2006, 79(1): 35-67.
[7] CHENG X, LIU S, SUN X, et al. Combating emerging financial risks in the big data era: a perspective review[J]. Fundamental Research, 2021, 1(5): 595-606.
[8] CHEN F, WANG Y, WANG B, et al. Graph representation learning: a survey[J]. APSIPA Transactions on Signal and Information Processing, 2020, 9: e15.
[9] GAO J, RIBEIRO B. On the equivalence between temporal and static graph representations for observational predictions[J]. arXiv:2103.07016, 2021.
[10] DONG Y, CHAWLA N, SWAMI A. Metapath2vec: scalable representation learning for heterogeneous networks[C]//Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2017: 135-144.
[11] 袁立宁, 李欣, 王晓冬, 等. 图嵌入模型综述[J]. 计算机科学与探索, 2022, 16(1): 59-87.
YUAN L N, LI X, WANG X D, et al. Graph embedding models: a survey[J]. Journal of Frontiers of Compute Science and Technology, 2022, 16(1): 59-87.
[12] BARROS C, MENDONCA M, VIEIRA A, et al. A survey on embedding dynamic graphs[J]. ACM Computing Surveys, 2021, 55(1): 1-37.
[13] YIN Y, JI L, ZHANG J, et al. DHNE: network representation learning method for dynamic heterogeneous networks[J]. IEEE Access, 2019, 7: 134782-134792.
[14] ZHOU L, YANG Y, REN X, et al. Dynamic network embedding by modeling triadic closure process[C]//Proceedings of the AAAI Conference on Artificial Intelligence, 2018.
[15] WANG X, LU Y, SHI C, et al. Dynamic heterogeneous information network embedding with meta-path based proximity[J]. IEEE Transactions on Knowledge and Data Engineering, 2020, 34(3): 1117-1132.
[16] GUO X, ZHOU B, SKIENA S. Subset node representation learning over large dynamic graphs[C]//Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 2021: 516-526.
[17] XU M. Understanding graph embedding methods and their applications[J]. SIAM Review, 2021, 63(4): 825-853.
[18] GOYAL P, FERRARA E. Graph embedding techniques, applications, and performance: a survey[J]. Knowledge-Based Systems, 2018, 151: 78-94.
[19] CAI H, ZHENG V, CHANG K. A comprehensive survey of graph embedding: problems, techniques, and applications[J]. IEEE Transactions on Knowledge and Data Engineering, 2018, 30(9): 1616-1637.
[20] 罗世杰, 吕文韬, 李凡, 等. 融合拓扑和属性的动态网络链路预测方法[J]. 计算机工程与应用, 2023, 59(5): 122-130.
LUO S J, LYU W T, LI F, et al. Dynamic network link prediction method for fusion topology and attributes[J]. Computer Engineering and Applications, 2023, 59(5): 122-130.
[21] GAO C, ZHU J, ZHANG F, et al. A novel representation learning for dynamic graphs based on graph convolutional networks[J]. IEEE Transactions on Cybernetics, 2022, 53(6): 3599-3612.
[22] WANG W, XIA F, NIE H, et al. Vehicle trajectory clustering based on dynamic representation learning of internet of vehicles[J]. IEEE Transactions on Intelligent Transportation Systems, 2020, 22(6): 3567-3576.
[23] GAO H, LV C, ZHANG T, et al. A structure constraint matrix factorization framework for human behavior segmentation[J]. IEEE Transactions on Cybernetics, 2021, 52(12): 12978-12988.
[24] 卢志刚, 陈倩. 动态供应链网络中企业合作关系的链路预测[J]. 计算机工程与应用, 2022, 58(2): 265-273.
LU Z G, CHEN Q. Link prediction of enterprise cooperation relationship in dynamic supply chain network[J]. Computer Engineering and Applications, 2022, 58(2): 265-273.
[25] TAHERI A, GIMPEL K, BERGER-WOLF T. Learning to represent the evolution of dynamic graphs with recurrent models[C]//Companion Proceedings of the 2019 World Wide Web Conference, 2019: 301-307.
[26] DU L, WANG Y, SONG G, et al. Dynamic network embedding: an extended approach for skip-gram based network embedding[C]//Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence, 2018: 2086-2092.
[27] XIE Y, YU B, LV S, et al. A survey on heterogeneous network representation learning[J]. Pattern Recognition, 2021, 116: 107936.
[28] BIAN R, KOH Y, DOBBIE G, et al. Network embedding and change modeling in dynamic heterogeneous networks[C]//Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval, 2019: 861-864.