Vehicle insurance fraud detection method based on improved graph attention network

doi:10.11772/j.issn.1001-9081.2025020151

Journal of Computer Applications ›› 2026, Vol. 46 ›› Issue (2): 437-444.DOI: 10.11772/j.issn.1001-9081.2025020151

• Artificial intelligence • Previous Articles

Vehicle insurance fraud detection method based on improved graph attention network

Jinjiao LIN¹, Canshun ZHANG¹, Shuya CHEN¹, Tianxin WANG¹, Jian LIAN²(), Yonghui XU³

^1.School of Management Science and Engineering，Shandong University of Finance and Economics，Jinan Shandong 250014，China
^2.School of Intelligent Engineering，Shandong Management University，Jinan Shandong 250000，China
^3.Joint SDU-NTU Centre for Artificial Intelligence Research，Shandong University，Jinan Shandong 250101，China

Received:2025-02-19 Revised:2025-05-10 Accepted:2025-05-13 Online:2025-05-16 Published:2026-02-10
Contact: Jian LIAN
About author:LIN Jinjiao， born in 1978， Ph. D.， professor. Her research interests include financial big data processing， artificial intelligence， deep learning.
ZHANG Canshun， born in 2002， M. S. candidate. His research interests include financial big data analysis and processing， deep learning.
CHEN Shuya， born in 2004. Her research interests include financial big data analysis and processing， deep learning.
WANG Tianxin， born in 2005. His research interests include financial big data analysis and processing， deep learning.
LIAN Jian， born in 1981， Ph. D.， professor. His research interests include image processing， deep learning， artificial intelligence. Email:lianjianlianjian@163.com
XU Yonghui， born in 1986， Ph. D.， professor. His research interests include knowledge graph， trusted artificial intelligence.
Supported by:
National Natural Science Foundation of China(62202279);Construction Project of “Digital Intelligence Empowering · Cross Integration” Undergraduate Course in Shandong University of Finance and Economics （Financial Data Collection and Processing）

基于改进图注意力网络的车险欺诈检测方法

林金娇¹, 张灿舜¹, 陈淑娅¹, 王天鑫¹, 连剑²(), 徐庸辉³

^1.山东财经大学管理科学与工程学院，济南 250014
^2.山东管理学院智能工程学院，济南 250000
^3.山东大学—南洋理工大学人工智能国际联合研究院，济南 250101

通讯作者: 连剑
作者简介:林金娇（1978—），女，浙江永康人，教授，博士，CCF会员，主要研究方向：金融大数据处理、人工智能、深度学习
张灿舜（2002—），男，山东泰安人，硕士研究生，CCF会员，主要研究方向：金融大数据分析与处理、深度学习
陈淑娅（2004—），女，湖北宜昌人，主要研究方向：金融大数据分析与处理、深度学习
王天鑫（2005—），男，山东济南人，主要研究方向：金融大数据分析与处理、深度学习
连剑（1981—），男，山东荣成人，教授，博士，CCF会员，主要研究方向：图像处理、深度学习、人工智能 Email:lianjianlianjian@163.com
徐庸辉（1986—），男，山东济南人，教授，博士，CCF会员，主要研究方向：知识图谱、可信人工智能。
基金资助:
国家自然科学基金资助项目(62202279);山东财经大学“数智赋能·交叉融合”本科生课程立项建设项目（金融数据采集与处理）

Abstract

Abstract:

Aiming at the problem that the existing vehicle insurance fraud detection methods ignore complex correlation in the data， a vehicle insurance fraud detection method based on improved graph attention network was proposed. This method enhances the ability to capture complex correlation in the data through collaborative design of dynamic attention mechanism and serialized global modeling. Firstly， each case of vehicle insurance fraud was abstracted as a node of graph structure. Secondly， the similarity between multiple attributes such as time， age， and amount of the nodes was calculated by K-Nearest Neighbor （KNN） algorithm， so as to construct the complex correlation between the cases. Thirdly， the graph data of the cases was input into GATv2（dynamic Graph ATtention network）， and local features of the adjacent nodes were aggregated by allocating node weights dynamically， thereby obtaining new feature representation of each case node. Fourthly， Transformer was introduced to serialize the graph structure output of GATv2. Finally the fusion module was used to perform nonlinear integration expression on the final features， so as to obtain the classification results of the case nodes. Experimental results show that compared with the baseline methods， the proposed method has the accuracy on the two datasets improved by at least 1.11 and 1.34 percentage points， respectively， and the False Positive Rate （FPR） of as low as 0.035% on the insurance company dataset， which provides a new technical solution for improving the accuracy and efficiency of vehicle insurance fraud detection.

Key words: vehicle insurance fraud detection, Graph Neural Network (GNN), graph attention mechanism, serialization, deep learning

摘要：

针对现有车险欺诈检测方法忽略数据中复杂关联关系的问题，提出一种基于改进图注意力网络的车险欺诈检测方法。该方法通过动态注意力机制与序列化全局建模的协同设计，增强数据中复杂关联关系的捕捉能力。首先，将车险欺诈的每一起案件抽象为图结构的节点；其次，通过K近邻（KNN）算法计算节点的时间、年龄以及金额等多重属性间的相似性，从而构建案件之间的复杂关联关系；再次，将案件的图数据输入GATv2（dynamic Graph ATtention network），并动态分配节点权重以聚合邻接节点的局部特征，从而得到每个案件节点的新特征表示；继次，引入Transformer对GATv2的图结构输出进行序列化处理；最后，由融合模块对最终特征进行非线性整合表达，从而得到案件节点的分类结果。实验结果表明：所提方法在两个数据集上的准确率较基线方法分别至少提升了1.11和1.34个百分点，而在保险公司数据集上的误警率（FPR）低至0.035%，为提高车险欺诈检测的准确性与效率提供了新的技术方案。

关键词: 车险欺诈检测, 图神经网络, 图注意力机制, 序列化, 深度学习

CLC Number:

TP183

Jinjiao LIN, Canshun ZHANG, Shuya CHEN, Tianxin WANG, Jian LIAN, Yonghui XU. Vehicle insurance fraud detection method based on improved graph attention network[J]. Journal of Computer Applications, 2026, 46(2): 437-444.

林金娇, 张灿舜, 陈淑娅, 王天鑫, 连剑, 徐庸辉. 基于改进图注意力网络的车险欺诈检测方法[J]. 《计算机应用》唯一官方网站, 2026, 46(2): 437-444.

Figures/Tables 8

References 36

[1]	李炜卓，卢冰洁，杨骏铭，等. 基于网络表示学习的车险欺诈溯因分析研究［J］. 计算机科学， 2023， 50（2）： 300-309.
	LI W Z， LU B J， YANG J M， et al. Study on abductive analysis of auto insurance fraud based on network representation learning［J］. Computer Science， 2023， 50（2）： 300-309.
[2]	喻炜，冯根福，张文珺. 机动车辆保险欺诈检测系统及团伙识别研究［J］. 保险研究， 2017（2）： 63-73.
	YU W， FENG G F， ZHANG W J. A research on fraud detection system and gang identification of motor vehicle insurance［J］. Insurance Studies， 2017（2）： 63-73.
[3]	ŠUBELJ L， ŠURLAN S， BAJEC M. An expert system for detecting automobile insurance fraud using social network analysis［J］. Expert Systems with Applications， 2011， 38（1）： 1039-1052.
[4]	刁莉，王宁. 基于X12-LSTM模型的保费收入预测研究［J］. 计算机科学， 2020， 47（6A）： 512-516.
	DIAO L， WANG N. Research on premium income forecast based on X12-LSTM model［J］. Computer Science， 2020， 47（6A）： 512-516.
[5]	BAUMANN M. Improving a rule-based fraud detection system with classification based on association rule mining［EB/OL］. ［2024-10-16］..
[6]	VIAENE S， DEDENE G， DERRIG R A. Auto claim fraud detection using Bayesian learning neural networks［J］. Expert Systems with Applications， 2005， 29（3）： 653-666.
[7]	PHUA C， LEE V， SMITH K， et al. A comprehensive survey of data mining-based fraud detection research［EB/OL］. ［2025-01-15］..
[8]	肖国庆，李雪琪，陈玥丹，等. 大规模图神经网络研究综述［J］. 计算机学报， 2024， 47（1）： 148-171.
	XIAO G Q， LI X Q， CHEN Y D， et al. A survey of large-scale graph neural networks［J］. Chinese Journal of Computers， 2024， 47（1）： 148-171.
[9]	江旭晖，沈英汉，李子健，等. 社交知识图谱研究综述［J］. 计算机学报， 2023， 46（2）： 304-330.
	JIANG X H， SHEN Y H， LI Z J， et al. A survey of social knowledge graph［J］. Chinese Journal of Computers， 2023， 46（2）： 304-330.
[10]	钱忠胜，赵畅，俞情媛，等. 结合注意力CNN与GNN的信息融合推荐方法［J］. 软件学报， 2023， 34（5）： 2317-2336.
	QIAN Z S， ZHAO C， YU Q Y， et al. Information fusion recommendation approach combining attention CNN and GNN［J］. Journal of Software， 2023， 34（5）： 2317-2336.
[11]	LI Y， ZHANG G， WANG P， et al. Graph neural networks for biomedical data： a review［J］. Current Bioinformatics， 2022， 17（6）： 483-492.
[12]	BRODY S， ALON U， YAHAV E. How attentive are graph attention networks？［EB/OL］. ［2025-01-16］..
[13]	ZHAO Y， ZHANG J， ZONG C. Transformer： a general framework from machine translation to others［J］. Machine Intelligence Research， 2023， 20（4）： 514-538.
[14]	GILLIOZ A， CASAS J， MUGELLINI E， et al. Overview of the Transformer-based models for NLP tasks［C］// Proceedings of the 15th Federated Conference on Computer Science and Information Systems. Piscataway： IEEE， 2020： 179-183.
[15]	卢冰洁，李炜卓，那崇宁，等. 机器学习模型在车险欺诈检测的研究进展［J］. 计算机工程与应用， 2022， 58（5）： 34-49.
	LU B J， LI W Z， NA C N， et al. Survey of auto insurance fraud detection with machine learning models［J］. Computer Engineering and Applications， 2022， 58（5）： 34-49.
[16]	陈凯，李斌杰. 主流机器学习方法识别车险欺诈效果的比较研究［J］. 保险研究， 2022（12）： 90-102.
	CHEN K， LI B J. A comparative study on the effectiveness of machine learning methods in auto insurance fraud identification［J］. Insurance Studies， 2022（12）： 90-102.
[17]	陈思迎. 大数据背景下机动车辆保险欺诈风险及其防范研究［D］. 成都：西南财经大学， 2019： 6-10.
	CHEN S Y. Research on risk and prevention of automobile insurance fraud in the background of big data［D］. Chengdu： Southwest University of Finance and Economics， 2019： 6-10.
[18]	TONGESAI M， MBIZO G， ZVAREVASHE K. Insurance fraud detection using machine learning［C］// Proceedings of the 1st Zimbabwe Conference of Information and Communication Technologies. Piscataway： IEEE， 2022： 1-6.
[19]	DEBENER J， HENINKE V， KRIEBEL J. Detecting insurance fraud using supervised and unsupervised machine learning［J］. Journal of Risk and Insurance， 2023， 90（3）： 743-768.
[20]	HANCOCK J， BAUDER R A， WANG H， et al. Explainable machine learning models for Medicare fraud detection［J］. Journal of Big Data， 2023， 10： No.154.
[21]	XIA H， ZHOU Y， ZHANG Z. Auto insurance fraud identification based on a CNN-LSTM fusion deep learning model［J］. International Journal of Ad Hoc and Ubiquitous Computing， 2022， 39（1/2）： 37-45.
[22]	MING R， ABDELRAHMAN O， INNAB N， et al. Enhancing fraud detection in auto insurance and credit card transactions： a novel approach integrating CNNs and machine learning algorithms［J］. PeerJ Computer Science， 2024， 10： No.e2088.
[23]	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. New York： ACM， 2021： 3168-3177.
[24]	ZHANG G， WU J， YANG J， et al. FRAUDRE： fraud detection dual-resistant to graph inconsistency and imbalance［C］// Proceedings of the 2021 IEEE International Conference on Data Mining. Piscataway： IEEE， 2021： 867-876.
[25]	LIU Y， SUN Z， ZHANG W. Improving fraud detection via hierarchical attention-based Graph Neural Network［J］. Journal of Information Security and Applications， 2023， 72： No.103399.
[26]	陈卓，朱淼，杜军威. 基于多视角图神经网络的欺诈检测算法［J］. 通信学报， 2022， 43（11）： 225-232.
	CHEN Z， ZHU M， DU J W. Multi-view graph neural network for fraud detection algorithm［J］. Journal on Communications， 2022， 43（11）： 225-232.
[27]	HILAL W， GADSDEN S A， YAWNEY J. Financial fraud： a review of anomaly detection techniques and recent advances［J］. Expert Systems with Applications， 2022， 193： No.116429.
[28]	WEI S， LEE S. Financial anti-fraud based on dual-channel graph attention network［J］. Journal of Theoretical and Applied Electronic Commerce Research， 2024， 19（1）： 297-314.
[29]	潘怡君，梁变，张泷，等. 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.
	PAN Y J， LIANG B， ZHANG L， et al. Dynamic heterogeneous network representation learning for fraud detection in auto insurance［J］. Computer Engineering and Applications， 2024， 60（24）： 322-330.
[30]	YANG J， CHEN K， DING K， et al. Auto insurance fraud detection with multimodal learning［J］. Data Intelligence， 2023， 5（2）： 388-412.
[31]	SHAH B. Auto insurance claims data［DS/OL］. ［2025-01-18］..
[32]	KAPOOR K. Vehicle insurance fraud detection［DS/OL］. ［2025-01-18］..
[33]	PAVLYSHENKO B. Using stacking approaches for machine learning models［C］// Proceedings of the IEEE 2nd International Conference on Data Stream Mining and Processing. Piscataway： IEEE， 2018： 255-258.
[34]	KUO P C， CHOU Y T， LI K Y， et al. GNN-LSTM-based fusion model for structural dynamic responses prediction［J］. Engineering Structures， 2024， 306： No.117733.
[35]	WU H， XU J， WANG J， et al. Autoformer： decomposition Transformers with auto-correlation for long-term series forecasting［C］// Proceedings of the 35th International Conference on Neural Information Processing Systems. Red Hook： Curran Associates Inc.， 2021： 22419-22430.
[36]	ZHOU H， ZHANG S， PENG J， et al. Informer： beyond efficient Transformer for long sequence time-series forecasting［C］// Proceedings of the 35th AAAI Conference on Artificial Intelligence. Palo Alto： AAAI Press， 2021： 11106-11115.

数据集	正常索赔	欺诈索赔
七州数据集	753	247
保险公司数据集	14 497	923

数据集	正常索赔	欺诈索赔
七州数据集	753	247
保险公司数据集	14 497	923

方法	七州数据集				保险公司数据集
方法	精确率/%	召回率/%	准确率/%	ADTS/ms	精确率/%	召回率/%	准确率/%	ADTS/ms
Stacking^［33］	84.34	84.39	84.37	1.74	97.97	97.64	97.83	1.78
GATv2^［12］	84.45	82.67	83.47	2.17	98.87	98.02	98.34	2.08
LSTM+GNN^［34］	87.42	87.21	87.17	1.01	96.74	95.66	96.12	1.21
Autoformer^［35］	87.67	87.88	87.96	1.69	97.64	97.55	97.61	1.65
Informer^［36］	87.76	87.95	87.90	2.87	96.67	95.78	96.25	3.07
本文方法	89.40	88.82	89.07	3.20	99.96	99.40	99.68	3.26

方法	七州数据集				保险公司数据集
方法	精确率/%	召回率/%	准确率/%	ADTS/ms	精确率/%	召回率/%	准确率/%	ADTS/ms
Stacking^［33］	84.34	84.39	84.37	1.74	97.97	97.64	97.83	1.78
GATv2^［12］	84.45	82.67	83.47	2.17	98.87	98.02	98.34	2.08
LSTM+GNN^［34］	87.42	87.21	87.17	1.01	96.74	95.66	96.12	1.21
Autoformer^［35］	87.67	87.88	87.96	1.69	97.64	97.55	97.61	1.65
Informer^［36］	87.76	87.95	87.90	2.87	96.67	95.78	96.25	3.07
本文方法	89.40	88.82	89.07	3.20	99.96	99.40	99.68	3.26

实验组	准确率/%
实验组	七州数据集	保险公司数据集
完整模型	89.07	99.68
仅GATv2	83.47	98.34
仅Transformer	82.75	96.23
无融合模块	87.42	98.45

Vehicle insurance fraud detection method based on improved graph attention network

基于改进图注意力网络的车险欺诈检测方法

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 8

References 36

Related Articles 15

Recommended Articles

Metrics

[1]	Haoqian JIANG, Dong ZHANG, Guanyu LI, Heng CHEN. SetaCRS： Conversational recommender system with structure-enhanced hierarchical task-oriented prompting strategy [J]. Journal of Computer Applications, 2026, 46(2): 368-377.
[2]	Zeyi GUO, Fenglian LI, Lichun XU. Double decision mechanism-based deep symbolic regression algorithm [J]. Journal of Computer Applications, 2026, 46(2): 406-415.
[3]	Wen LI, Kairong LI, Kai YANG. Subgraph-aware contrastive learning with data augmentation [J]. Journal of Computer Applications, 2026, 46(1): 1-9.
[4]	Xiaoyong BIAN, Peiyang YUAN, Qiren HU. Dual-coding space-frequency mixing method for infrared small target detection [J]. Journal of Computer Applications, 2026, 46(1): 252-259.
[5]	Fan HE, Li LI, Zhongxu YUAN, Xiu YANG, Dongxuan HAN. Knowledge tracking model based on concept association memory network with graph attention [J]. Journal of Computer Applications, 2026, 46(1): 43-51.
[6]	Zhihui ZAN, Yajing WANG, Ke LI, Zhixiang YANG, Guangyu YANG. Multi-feature fusion speech emotion recognition method based on SAA-CNN-BiLSTM network [J]. Journal of Computer Applications, 2026, 46(1): 69-76.
[7]	Chao LIU, Yanhua YU. Knowledge-aware recommendation model combining denoising strategy and multi-view contrastive learning [J]. Journal of Computer Applications, 2025, 45(9): 2827-2837.
[8]	Hongjun ZHANG, Gaojun PAN, Hao YE, Yubin LU, Yiheng MIAO. Multi-source heterogeneous data analysis method combining deep learning and tensor decomposition [J]. Journal of Computer Applications, 2025, 45(9): 2838-2847.
[9]	Jin LI, Liqun LIU. SAR and visible image fusion based on residual Swin Transformer [J]. Journal of Computer Applications, 2025, 45(9): 2949-2956.
[10]	Bing YIN, Zhenhua LING, Yin LIN, Changfeng XI, Ying LIU. Emotion recognition method compatible with missing modal reasoning [J]. Journal of Computer Applications, 2025, 45(9): 2764-2772.
[11]	Weigang LI, Jiale SHAO, Zhiqiang TIAN. Point cloud classification and segmentation network based on dual attention mechanism and multi-scale fusion [J]. Journal of Computer Applications, 2025, 45(9): 3003-3010.
[12]	Zhixiong XU, Bo LI, Xiaoyong BIAN, Qiren HU. Adversarial sample embedded attention U-Net for 3D medical image segmentation [J]. Journal of Computer Applications, 2025, 45(9): 3011-3016.
[13]	Panfeng JING, Yudong LIANG, Chaowei LI, Junru GUO, Jinyu GUO. Semi-supervised image dehazing algorithm based on teacher-student learning [J]. Journal of Computer Applications, 2025, 45(9): 2975-2983.
[14]	Lina GE, Mingyu WANG, Lei TIAN. Review of research on efficiency of federated learning [J]. Journal of Computer Applications, 2025, 45(8): 2387-2398.
[15]	Yi WANG, Yinglong MA. Multi-task social item recommendation method based on dynamic adaptive generation of item graph [J]. Journal of Computer Applications, 2025, 45(8): 2592-2599.