机器学习模型在车险欺诈检测的研究进展

doi:10.3778/j.issn.1002-8331.2109-0312

摘要/Abstract

摘要： 随着保险行业的蓬勃发展，保险欺诈问题也显得日趋严重。车险欺诈一直是保险欺诈的“重灾区”，对保险行业的发展至关重要。因此，车险欺诈检测技术一直是国内外学者研究的热点问题。鉴于我国在机动车辆保险欺诈检测技术方相对滞后，而国外的研究成果又较少对我国车险业务数据进行有效建模与分析，首次针对机器学习模型应用在车险欺诈检测的研究工作进行了文献调研，对二十多年来的研究工作进行系统化的归纳与总结。通过引入车险欺诈流程的简介，对专家系统与智能理赔系统在车险欺诈检测的流程进行了叙述；依次从国外和国内的角度介绍了机器学习模型应用在车险欺诈检测的具体研究进展，并进行了宏观的对比；基于国内某车险公司提供近5年来高质量的车险数据选取具有代表性的机器学习模型进行建模，并进行了全面的测试与分析；探讨了车险欺诈检测技术未来的研究方向。

关键词: 汽车保险欺诈, 机器学习, 深度学习, 数据不均衡, 保险监管

Abstract: With the vigorous development of insurance industry, the problem of insurance fraud has become increasingly serious. Auto insurance fraud has been the hardest “hit area” of insurance fraud, which is very important for the development of insurance industry. Therefore, auto insurance fraud detection technology has been a hot top for researchers. Considered that the fraud detection techniques of automobiles in China have been lagged, and research results in aboard are not enough to model and analyze China’s auto insurance business data, this paper presents the first work on the survey of the methods of machine learning applied in auto insurance fraud detection, and systematically summarizes the related works in the two decades. It firstly gives a brief introduction to the process of auto insurance fraud detection based on expert systems and intelligent claim settlement systems. Then, it describes the specific research progresses of machine learning models applied in auto insurance fraud detection at home and aboard. Moreover, with the high-quality auto insurance data provided by the insurance company in recent five years, it carries out a detailed comparison based on the representative models from machine learning and gives a comprehensive analysis. Finally, the future research direction of auto insurance fraud detection is discussed and outlooked.

Key words: auto insurance fraud, machine learning, deep learning, unbalanced data, insurance regulation

卢冰洁, 李炜卓, 那崇宁, 牛作尧, 陈奎. 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 Bingjie, LI Weizhuo, NA Chongning, NIU Zuoyao, CHEN Kui. Survey of Auto Insurance Fraud Detection with Machine Learning Models[J]. Computer Engineering and Applications, 2022, 58(5): 34-49.

参考文献

[1] 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 vehicle insurance[J].Insurance Studies，2017（2）：63-73.
[2] 车险反欺诈联合课题组.车险欺诈与反欺诈问题研究及监管建议[J].保险研究，2021（6）：3-10.
The Joint Research Team on Anti-Vehicle Frauds.A research on vehicle insurance fraud and anti-fraud and issues and regulatory suggestions[J].Insurance Studies，2021（6）：3-10.
[3] 张碧怡，肖宇谷，曾宇哲.车险定价中风险因子重要性测度的比较研究——基于集成学习方法和广义线性回归模型[J].保险研究，2019（10）：73-83.
ZHANG B Y，XIAO Y G，ZENG Y Z.A comparative study on measuring variable importance in auto insurance pricing—based on ensemble learning and generalized linear regression[J].Insurance Studies，2019（10）：73-83.
[4] 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.
[5] HANAFIZADEH P，PAYDAR N R.A data mining model for risk assessment and customer segmentation in the insurance industry[J].International Journal of Strategic Decision Sciences，2013，4（1）：51-77.
[6] KA??ELAN L，KA??ELAN V，NOVOVI?-BURI? M.A data mining approach for risk assessment in car insurance：evidence from Montenegro[J].International Journal of Business Intelligence Research，2014，5（3）：11-28.
[7] LI Y，YAN C，LIU W，et al.A principle component analysis-based random forest with the potential nearest neighbor method for automobile insurance fraud identification[J].Applied Soft Computing，2018，70：1000-1009.
[8] HE X，CHUA T S.Neural factorization machines for sparse predictive analytics[C]//Proceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval，2017：355-364.
[9] GUO J，LIU G，ZUO Y，et al.Learning sequential behavior representations for fraud detection[C]//2018 IEEE International Conference on Data Mining（ICDM），2018：127-136.
[10] WANG R，FU B，FU G，et al.Deep & cross network for ad click predictions[C]//Proceedings of the ADKDD’17，2017：1-7.
[11] SUBUDHI S，PANIGRAHI S.Use of optimized fuzzy C-means clustering and supervised classifiers for automobile insurance fraud detection[J].Journal of King Saud University-Computer and Information Sciences，2020，32（5）：568-575.
[12] MAJHI S K.Fuzzy clustering algorithm based on modified whale optimization algorithm for automobile insurance fraud detection[J].Evolutionary Intelligence，2021，14（1）：35-46.
[13] 庹国柱，段家喜.保险欺诈的博弈分析[J].首都经济贸易大学学报，1999（3）：51-54.
GUO G Z，DUAN J X.Game analysis of insurance fraud[J].Journal of Capital University of Economics and Business，1999（3）：51-54.
[14] 刘喜华，金加林.保险欺诈博弈与基于最优博弈策略的保险契约[J].系统工程理论与实践，2004（2）：19-24.
LIU X H，JIN J L.The insurance fraud game and insurance contract based on optimal game strategies[J].System Engineering Theory and Practice，2004（2）：19-24.
[15] 桂萍，胡庆为.国内外车险道德风险文献综述[J].保险研究，2011（6）：121-127.
GUI P，HU Q W.An overview on the domestic and foreign literature on moral risks related to motor insurance[J].Insurance Studies，2011（6）：121-127.
[16] 赵桂芹，吴洪.汽车保险市场中存在道德风险吗？——来自动态续保数据的分析[J].金融研究，2010（6）：175-188.
ZHAO G Q，WU H.Is there moral hazard in Chinese automobile insurance market？—evidence from dynamic renewal data[J].Journal of Financial Research，2010（6）：175-188.
[17] 汤俊，莫依雯.基于数据挖掘技术的车险反欺诈系统构建[J].上海保险，2013（11）：39-42.
TANG J，MO Y W.Construction of auto insurance anti-fraud system based on data mining technology[J].Shanghai Insurance，2013（11）：39-42.
[18] 王海巍.我国险企运营中道德风险甄别问题研究——以大数据Hadoop聚类分析技术为视角[J].保险研究，2016（2）：59-67.
WANG H W.A research on Chinese insurers’moral hazard screening in operation：from the big data Hadoop clustering analysis technology perspective[J].Insurance Studies，2016（2）：59-67.
[19] 闫春，李亚琪，孙海棠.基于蚁群算法优化随机森林模型的汽车保险欺诈识别研究[J].保险研究，2017（6）：114-127.
YAN C，LI Y Q，SUN H T.A research on automobile insurance fraud identification based on random forest model and ant colony optimization algorithm[J].Insurance Studies，2017（6）：114-127.
[20] 徐徐，王正祥，王牧群.基于深度学习技术的机动车辆保险欺诈识别模型与实证研究[J].上海保险，2019（8）：53-58.
XU X，WANG Z X，WANG M Q.The model and empiri-cal study of motor vehicle insurance fraud identification based on deep learning technology[J].Shanghai Insurance，2019（8）：53-58.
[21] PANIGRAHI S，PALKAR B.Comparative analysis on classification algorithms of auto-insurance fraud detection based on feature selection algorithms[J].Int J Comput Sci Eng，2018，6（9）：72-77.
[22] B?ASZCZY?SKI J，DE ALMEIDA FILHO A T，MATUSZYK A，et al.Auto loan fraud detection using dominance-based rough set approach versus machine learning methods[J].Expert Systems with Applications，2021，163：113740.
[23] HASSAN A K I，ABRAHAM A.Modeling insurance fraud detection using imbalanced data classification[M]//Advances in nature and biologically inspired computing.Cham：Springer，2016：117-127.
[24] PADHI S，PANIGRAHI S.Decision templates based ensemble classifiers for automobile insurance fraud detection[C]//2019 Global Conference for Advancement in Technology（GCAT），2019：1-5.
[25] ?UBELJ L，FURLAN ?，BAJEC M.An expert system for detecting automobile insurance fraud using social network analysis[J].Expert Systems with Applications，2011，38（1）：1039-1052.
[26] BERMúDEZ L，PéREZ J M，AYUSO M，et al.A Bayesian dichotomous model with asymmetric link for fraud in insurance[J].Insurance：Mathematics and Economics，2008，42（2）：779-786.
[27] BHOWMIK R.Detecting auto insurance fraud by data mining techniques[J].Journal of Emerging Trends in Computing and Information Sciences，2011，2（4）：156-162.
[28] YAN C，LI Y.The identification algorithm and model construction of automobile insurance fraud based on data mining[C]//2015 Fifth International Conference on Instrumentation and Measurement，Computer，Communication and Control（IMCCC），2015：1922-1928.
[29] MIHAELA D，D?NU?VASILE J.Modeling the frequency of auto insurance claims by means of Poisson and negative binomial models[J].Annals of the Alexandru Ioan Cuza University-Economics，2015，62（2）：151-168.
[30] NIAN K，ZHANG H，TAYAL A，et al.Auto insurance fraud detection using unsupervised spectral ranking for anomaly[J].The Journal of Finance and Data Science，2016，2（1）：58-75.
[31] BADRIYAH T，RAHMANIAH L，SYARIF I.Nearest neighbour and statistics method based for detecting fraud in auto insurance[C]//2018 International Conference on Applied Engineering（ICAE），2018：1-5.
[32] YAN C，LI Y，LIU W，et al.An artificial bee colony-based kernel ridge regression for automobile insurance fraud identification[J].Neurocomputing，2020，393：115-125.
[33] XU W，WANG S，ZHANG D，et al.Random rough subspace based neural network ensemble for insurance fraud detection[C]//2011 Fourth International Joint Conference on Computational Sciences and Optimization，2011：1276-1280.
[34] DHIEB N，GHAZZAI H，BESBES H，et al.A secure AI-driven architecture for automated insurance systems：fraud detection and risk measurement[J].IEEE Access，2020，8：58546-58558.
[35] LIANG C，LIU Z，LIU B，et al.Uncovering insurance fraud conspiracy with network learning[C]//Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval，2019：1181-1184.
[36] LIU Z，DOU Y，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.
[37] QU Y，CAI H，REN K，et al.Product-based neural networks for user response prediction[C]//2016 IEEE 16th International Conference on Data Mining（ICDM），2016：1149-1154.
[38] CHENG H T，KOC L，HARMSEN J，et al.Wide & deep learning for recommender systems[C]//Proceedings of the 1st Workshop on Deep Learning for Recommender Systems，2016：7-10.
[39] XIAO J，YE H，HE X，et al.Attentional factorization machines：learning the weight of feature interactions via attention networks[C]//Proceedings of the 26th International Joint Conference on Artificial Intelligence，2017：3119-3125.
[40] WANG Y，XU W.Leveraging deep learning with LDA-based text analytics to detect automobile insurance fraud[J].Decision Support Systems，2018，105：87-95.
[41] GUO H，TANG R，YE Y，et al.DeepFM：a factorization-machine based neural network for CTR prediction[C]//Proceedings of the 26th International Joint Conference on Artificial Intelligence，2017：1725-1731.
[42] GOMES C，JIN Z，YANG H.Insurance fraud detection with unsupervised deep learning[J].Journal of Risk and Insurance，2021，88（3）：591-624.
[43] 张连增，孙维伟.车险索赔概率影响因素的Logistic模型分析[J].保险研究，2012（7）：16-25.
ZHANG L Z，SUN W W.Analysis of factors affecting the probability of auto insurance claim occurrence with logistic model[J].Insurance Studies，2012（7）：16-25.
[44] 袁幕琴.保险欺诈与大数据时代下的防范对策[J].金融与经济，2015（7）：77-80.
YUAN M Q.Insurance fraud and countermeasures in the era of big data[J].Finance and Economy，2015（7）：77-80.
[45] 卢文龙.大数据与车险反欺诈模型构建[J].上海保险，2015（1）：27-30.
LU W L.Big data and auto insurance anti-fraud model construction[J].Shanghai Insurance，2015（1）：27-30.
[46] 张澄，赵晨.位置大数据在车辆保险风险管理中的应用[J].大数据，2016，2（5）：79-87.
ZHANG C，ZHAO C.Application of location-based big data in auto insurance risk control[J].Big Data Research，2016，2（5）：79-87.
[47] 叶明华.基于BP神经网络的保险欺诈识别研究——以中国机动车保险索赔为例[J].保险研究，2011（3）：79-86.
YE M H.Insurance fraud identification research based on BP neurological network—with China motor insurance claim as an example[J].Insurance Studies，2011（3）：79-86.
[48] 闫春，厉美璇，周潇.基于改进的遗传算法优化BP神经网络的车险欺诈识别模型[J].山东科技大学学报（自然科学版），2019，38（5）：72-80.
YAN C，LI M X，ZHOU X.Improved genetic algorithm for vehicle insurance fraud identification model based on BP neural network[J].Journal of Shandong University of Science and Technology（Natural Science），2019，38（5）：72-80.
[49] 赵长利，李方媛.汽车保险网络风险闭环控制优化模型及算法[J].系统工程，2016，34（3）：69-73.
ZHAO C L，LI F Y.An optimization model and algorithm of closed-loop control for the risks of automobile insurance network[J].System Engineering，2016，34（3）：69-73.
[50] CHEN W Y，LIU Y C，KIRA Z，et al.A closer look at few-shot classification[J].arXiv：1904.04232，2019.
[51] BENDRE N，MARíN H T，NAJAFIRAD P.Learning from few samples：a survey[J].arXiv：2007.15484，2020.
[52] 杨烁.基于图的车险欺诈可视化分析系统的设计与实现[D].北京：北京邮电大学，2019.
YANG S.The design and implementation of vehicle insurance fraud visual analysis system based on graph[D].Beijing：Beijing University of Posts and Telecommunications，2019.
[53] 漆桂林，高桓，吴天星.知识图谱研究进展[J].情报工程，2017，3（1）：4-25.
QI G L，GAO H，WU T X.The research advances of knowledge graph[J].Technology Intelligence Engineering，2017，3（1）：4-25.
[54] 胡志磊，靳小龙，陈剑赟，等.事件图谱的构建、推理与应用[J].大数据，2021，7（3）：80-96.
HU Z L，JIN X L，CHEN J Y，et al.Construction，reasoning and applications of event graphs[J].Big Data Research，2021，7（3）：80-96.
[55] FRANKENDAL J M，GHANBARISAIED S.Enabling innovation through an ecosystem：an exploratory case study[Z].2020.
[56] 曾春艳，严康，王志锋，等.深度学习模型可解释性研究综述[J].计算机工程与应用，2021，57（8）：1-9.
ZENG C Y，YAN K，WANG Z F，et al.Survey of interpretability research on deep learning models[J].Computer Engineering and Applications，2021，57（8）：1-9.
[57] 孙影影，贾振堂，朱昊宇.多模态深度学习综述[J].计算机工程与应用，2020，56（21）：1-10.
SUN Y Y，JIA Z T，ZHU H Y.Summary of multimodal deep learning[J].Computer Engineering and Applications，2020，56（21）：1-10.
[58] 杨杨，詹德川，姜远，等.可靠多模态学习综述[J].软件学报，2021，32（4）：1067-1081.
YANG Y，ZHAN D C，JIANG Y，et al.Reliable multi-modal learning：a survey[J].Journal of Software，2021，32（4）：1067-1081.
[59] 王雪鹏，刘康，何世柱，等.基于网络语义标签的多源知识库实体对齐算法[J].计算机学报，2017，40（3）：701-711.
WANG X P，LIU K，HE S Z，et al.Multi-source knowledge bases entity alignment by leveraging semantic tags[J].Chinese Journal of Computers，2017，40（3）：701-711.
[60] 陈焕新，李爱萍.基于Fabric的快速公平合同签署协议[J]计算机工程与应用，2022，58（2）：116-122.
CHEN H X，LI A P.Fast and fair contract signing protocol for fabric[J].Computer Engineering and Applications，2022，58（2）：116-122.
[61] 王健宗，孔令炜，黄章成，等.联邦学习算法综述[J].大数据，2020，6（6）：64-82.
WANG J Z，KONG L W，HUANG Z C，et al.Research review of federated learning algorithms[J].Big Data Research，2020，6（6）：64-82.
[62] 刘俊旭，孟小峰.机器学习的隐私保护研究综述[J].计算机研究与发展，2020，57（2）：346-362.
LIU J X，MENG X F.Survey on privacy-preserving machine learning[J].Journal of Computer Research and Development，2020，57（2）：346-362.