Generative hybrid models for fraud detection in auto insurance with a comparative analysis of VAE, GAN, and diffusion approaches

Fraud claim detection in auto insurance remains a vital yet complex challenge, mainly due to imbalanced data sets, non-linear feature interactions, and the necessity for explicable predictions. While traditional Machine Learning (ML) approaches show promise, they frequently struggle from poor genera...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	Discover Artificial Intelligence Ročník 5; číslo 1; s. 313 - 23
Hlavní autoři:	Bekkaye, Chadia, Oukhouya, Hassan, Zari, Tarek, Guerbaz, Raby, El Bouanani, Hicham
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	Cham Springer International Publishing 01.12.2025 Springer Nature B.V Springer
Témata:	Accuracy Algorithms Artificial Intelligence Auto insurance Automobile insurance Classification Comparative analysis Computer Science Datasets Engineering Feature selection Fraud detection Fraud prevention Generative hybrid models Health insurance Insurance claims Insurance fraud Isolation forest Machine leaning algorithms Medical research Methods Neural networks Oversampling methods Support vector machines Survival analysis Classification metrics Isolation forest Performance comparison Oversampling methods Probabilistic calibration Auto insurance Machine leaning algorithms Fraud detection Model stability Generative hybrid models
ISSN:	2731-0809, 2731-0809
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Abstract	Fraud claim detection in auto insurance remains a vital yet complex challenge, mainly due to imbalanced data sets, non-linear feature interactions, and the necessity for explicable predictions. While traditional Machine Learning (ML) approaches show promise, they frequently struggle from poor generalization, limited interpretability, and inadequate treatment of rare fraudulent cases. The present paper proposes a new hybrid approach involving generative models —namely Variational AutoEncoders (VAEs), Generative Adversarial Networks (GANs), and Diffusion Models (DMs)—with an ensemble of classifiers including eXtreme Gradient Boosting (XGBoost), Random Forest (RF), and Light Gradient Boosting (Light GBM), coupled with Isolation Forest (IF) for anomaly detection and oversampling-based techniques (SMOTE and ADASYN) to ameliorate class balance. In total, 18 hybrid combinations were developed and evaluated across classification performance (AUC-ROC, Accuracy, Precision, Recall, F1-score), probabilistic calibration (Brier Score and Log loss), and stochastic stability (Monte Carlo Variance and Bootstrap Variance). The experimental findings—backed up by graphical analysis based on radar plots, ROC curves, 3D metric visualization, and SHAP explainability—confirm that DM coupled with XGBoost and SMOTE (DM_XGBoost_SMOTE) and DM with Light GBM and SMOTE (DM_Light GBM_SMOTE) outperform alternative combinations. In particular, DM_XGBoost_SMOTE achieves a well balanced compromise between accuracy, confidence calibration, and robustness. This work underlines the efficiency of Diffusion-based hybrid models in fraud detection and opens the way for their implementation in high-risk, real-world insurance environments.
AbstractList	Abstract Fraud claim detection in auto insurance remains a vital yet complex challenge, mainly due to imbalanced data sets, non-linear feature interactions, and the necessity for explicable predictions. While traditional Machine Learning (ML) approaches show promise, they frequently struggle from poor generalization, limited interpretability, and inadequate treatment of rare fraudulent cases. The present paper proposes a new hybrid approach involving generative models —namely Variational AutoEncoders (VAEs), Generative Adversarial Networks (GANs), and Diffusion Models (DMs)—with an ensemble of classifiers including eXtreme Gradient Boosting (XGBoost), Random Forest (RF), and Light Gradient Boosting (Light GBM), coupled with Isolation Forest (IF) for anomaly detection and oversampling-based techniques (SMOTE and ADASYN) to ameliorate class balance. In total, 18 hybrid combinations were developed and evaluated across classification performance (AUC-ROC, Accuracy, Precision, Recall, F1-score), probabilistic calibration (Brier Score and Log loss), and stochastic stability (Monte Carlo Variance and Bootstrap Variance). The experimental findings—backed up by graphical analysis based on radar plots, ROC curves, 3D metric visualization, and SHAP explainability—confirm that DM coupled with XGBoost and SMOTE (DM_XGBoost_SMOTE) and DM with Light GBM and SMOTE (DM_Light GBM_SMOTE) outperform alternative combinations. In particular, DM_XGBoost_SMOTE achieves a well balanced compromise between accuracy, confidence calibration, and robustness. This work underlines the efficiency of Diffusion-based hybrid models in fraud detection and opens the way for their implementation in high-risk, real-world insurance environments. Fraud claim detection in auto insurance remains a vital yet complex challenge, mainly due to imbalanced data sets, non-linear feature interactions, and the necessity for explicable predictions. While traditional Machine Learning (ML) approaches show promise, they frequently struggle from poor generalization, limited interpretability, and inadequate treatment of rare fraudulent cases. The present paper proposes a new hybrid approach involving generative models —namely Variational AutoEncoders (VAEs), Generative Adversarial Networks (GANs), and Diffusion Models (DMs)—with an ensemble of classifiers including eXtreme Gradient Boosting (XGBoost), Random Forest (RF), and Light Gradient Boosting (Light GBM), coupled with Isolation Forest (IF) for anomaly detection and oversampling-based techniques (SMOTE and ADASYN) to ameliorate class balance. In total, 18 hybrid combinations were developed and evaluated across classification performance (AUC-ROC, Accuracy, Precision, Recall, F1-score), probabilistic calibration (Brier Score and Log loss), and stochastic stability (Monte Carlo Variance and Bootstrap Variance). The experimental findings—backed up by graphical analysis based on radar plots, ROC curves, 3D metric visualization, and SHAP explainability—confirm that DM coupled with XGBoost and SMOTE (DM_XGBoost_SMOTE) and DM with Light GBM and SMOTE (DM_Light GBM_SMOTE) outperform alternative combinations. In particular, DM_XGBoost_SMOTE achieves a well balanced compromise between accuracy, confidence calibration, and robustness. This work underlines the efficiency of Diffusion-based hybrid models in fraud detection and opens the way for their implementation in high-risk, real-world insurance environments.
ArticleNumber	313
Author	El Bouanani, Hicham Zari, Tarek Guerbaz, Raby Bekkaye, Chadia Oukhouya, Hassan
Author_xml	– sequence: 1 givenname: Chadia orcidid: 0009-0009-9610-8000 surname: Bekkaye fullname: Bekkaye, Chadia email: chadiabek2024@gmail.com organization: MAEGE Laboratory, Department of Statistics and Applied Mathematics, Hassan II University – sequence: 2 givenname: Hassan orcidid: 0000-0002-5058-2008 surname: Oukhouya fullname: Oukhouya, Hassan email: oukhouya.hassan@ump.ac.ma organization: MAEGE Laboratory, Department of Statistics and Applied Mathematics, Hassan II University, LaMSD, Team of MSASE, Department of Economics, Mohammed First University – sequence: 3 givenname: Tarek orcidid: 0000-0002-3549-128X surname: Zari fullname: Zari, Tarek organization: MAEGE Laboratory, Department of Statistics and Applied Mathematics, Hassan II University – sequence: 4 givenname: Raby orcidid: 0009-0009-7303-6008 surname: Guerbaz fullname: Guerbaz, Raby organization: MAEGE Laboratory, Department of Statistics and Applied Mathematics, Hassan II University – sequence: 5 givenname: Hicham orcidid: 0009-0002-5211-9831 surname: El Bouanani fullname: El Bouanani, Hicham organization: MAEGE Laboratory, Department of Statistics and Applied Mathematics, Hassan II University
BookMark	eNp9kU9v1DAQxSNUJErpF-BkiWsD43-xc1xVZalUwQW4WpN43PVqN17sBLS3fnTcbgWcOM1o9N5v9PReN2dTmqhp3nJ4zwHMh6IU72QLQrcA2qhWv2jOhZG8BQv92T_7q-aylC0ACKuN7OG8eVjTRBnn-JPY5jjk6Nk-edoVFlJmIePimaeZxjmmicWJ4TKnOsuScRqJ_YrzhiEb0_6AzxiccHcssbAU2PfVzRVbrz5f1WsFxRCW8gjCwyEnHDdU3jQvA-4KXT7Pi-bbx5uv15_auy_r2-vVXTvKTukWgwxgQk9DJwSOveq7YAYjAu_RDkpr9Nb3XqmRE3rTGVQCyEgPYQBrSF40tyeuT7h1hxz3mI8uYXRPh5TvHeY5jjtyyMlapTkHKRSRtjYICNDzQXA0na2sdydWDfFjoTK7bVpyjV2cFF0PRlndVZU4qcacSskU_nzl4B6Lc6fiXC3OPRXndDXJk6lU8XRP-S_6P67fuTqcvg
Cites_doi	10.1175/1520-0493(1950)078<0001:VOFEIT>2.0.CO;2 10.1007/978-3-030-66665-1_17 10.1109/ICDM.2008.17 10.1023/A:1010933404324 10.1016/j.asoc.2015.07.018 10.1016/j.dss.2017.11.001 10.6339/JDS.201207_10(3).0010 10.1145/2939672.2939785 10.1155/2021/6033860 10.7717/peerj-cs.2088 10.5391/IJFIS.2024.24.4.333 10.1088/2631-8695/ad5d51 10.1007/s11831-025-10271-2 10.1109/ACCESS.2020.2983300 10.1007/978-1-4899-4541-9 10.1007/978-3-319-98074-4 10.1007/978-3-030-96302-6_5 10.1109/TII.2024.3366991 10.14569/IJACSA.2020.0111054 10.1007/978-1-4757-4145-2 10.3390/a15050139 10.1109/IJCNN.2008.4633969 10.1111/jori.12359 10.1007/s10115-025-02377-7 10.1109/ACCESS.2024.3468993 10.1371/journal.pone.0325900 10.1016/j.ipm.2009.03.002 10.19139/soic-2310-5070-2655 10.1613/jair.953 10.1609/aaai.v30i1.9825 10.1016/j.ribaf.2022.101744 10.1111/1539-6975.00023 10.3390/math12020295 10.1109/TKDE.2004.1277822 10.11591/ijece.v14i1.pp911-921 10.1016/j.eswa.2005.04.030 10.3390/math11163605 10.1016/j.patrec.2005.10.010 10.1016/j.eswa.2020.114463 10.1080/17686733.2024.2349976 10.1088/2631-8695/ad76f9 10.1016/j.eswa.2023.120644 10.1186/s12859-023-05443-5 10.1016/S0957-4174(97)00045-6
ContentType	Journal Article
Copyright	The Author(s) 2025 The Author(s) 2025. This work is published under http://creativecommons.org/licenses/by-nc-nd/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
Copyright_xml	– notice: The Author(s) 2025 – notice: The Author(s) 2025. This work is published under http://creativecommons.org/licenses/by-nc-nd/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
DBID	C6C AAYXX CITATION 3V. 7XB 8FK 8G5 ABUWG AFKRA AZQEC BENPR CCPQU DWQXO GNUQQ GUQSH M2O MBDVC PHGZM PHGZT PIMPY PKEHL PQEST PQQKQ PQUKI PRINS Q9U DOA
DOI	10.1007/s44163-025-00574-5
DatabaseName	Springer Nature OA Free Journals CrossRef ProQuest Central (Corporate) ProQuest Central (purchase pre-March 2016) ProQuest Central (Alumni) (purchase pre-March 2016) Research Library (Alumni Edition) ProQuest Central (Alumni Edition) ProQuest Central UK/Ireland ProQuest Central Essentials ProQuest Central ProQuest One Community College ProQuest Central Korea ProQuest Central Student Research Library Prep Research Library Research Library (Corporate) ProQuest Central Premium ProQuest One Academic (New) Publicly Available Content Database ProQuest One Academic Middle East (New) ProQuest One Academic Eastern Edition (DO NOT USE) ProQuest One Academic (retired) ProQuest One Academic UKI Edition ProQuest Central China ProQuest Central Basic DOAJ Directory of Open Access Journals
DatabaseTitle	CrossRef Publicly Available Content Database Research Library Prep ProQuest Central Student ProQuest One Academic Middle East (New) ProQuest Central Basic ProQuest Central Essentials ProQuest One Academic Eastern Edition ProQuest Central (Alumni Edition) ProQuest One Community College Research Library (Alumni Edition) ProQuest Central China ProQuest Central ProQuest One Academic UKI Edition ProQuest Central Korea ProQuest Research Library ProQuest Central (New) ProQuest One Academic ProQuest One Academic (New) ProQuest Central (Alumni)
DatabaseTitleList	Publicly Available Content Database
Database_xml	– sequence: 1 dbid: DOA name: DOAJ Directory of Open Access Journals url: https://www.doaj.org/ sourceTypes: Open Website – sequence: 2 dbid: PIMPY name: Publicly Available Content Database url: http://search.proquest.com/publiccontent sourceTypes: Aggregation Database
DeliveryMethod	fulltext_linktorsrc
Discipline	Engineering Computer Science
EISSN	2731-0809
EndPage	23
ExternalDocumentID	oai_doaj_org_article_a1e8845110324ee588f20f091b21a768 10_1007_s44163_025_00574_5
GroupedDBID	0R~ 8G5 AAJSJ AAKKN AASML ABEEZ ABUWG ACACY ACULB ACVER AFFHD AFGXO AFKRA ALMA_UNASSIGNED_HOLDINGS ARCSS AZQEC BENPR C24 C6C CCPQU DWQXO EBLON EBS GNUQQ GROUPED_DOAJ GUQSH IAO ICD ITC M2O M~E OK1 PHGZM PHGZT PIMPY SOJ AAYXX CITATION 3V. 7XB 8FK MBDVC PKEHL PQEST PQQKQ PQUKI PRINS Q9U
ID	FETCH-LOGICAL-c3645-af3f07f9eb622ac9496f7b72f19a8b455ad8d9d44c1ead767a420e73d0fb087e3
IEDL.DBID	C24
ISSN	2731-0809
IngestDate	Mon Nov 10 19:22:51 EST 2025 Thu Nov 06 15:00:18 EST 2025 Thu Nov 13 04:29:53 EST 2025 Thu Nov 06 11:44:36 EST 2025
IsDoiOpenAccess	true
IsOpenAccess	true
IsPeerReviewed	true
IsScholarly	true
Issue	1
Keywords	Classification metrics Isolation forest Performance comparison Oversampling methods Probabilistic calibration Auto insurance Machine leaning algorithms Fraud detection Model stability Generative hybrid models
Language	English
LinkModel	DirectLink
MergedId	FETCHMERGED-LOGICAL-c3645-af3f07f9eb622ac9496f7b72f19a8b455ad8d9d44c1ead767a420e73d0fb087e3
Notes	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14
ORCID	0000-0002-5058-2008 0009-0009-9610-8000 0009-0009-7303-6008 0009-0002-5211-9831 0000-0002-3549-128X
OpenAccessLink	https://link.springer.com/10.1007/s44163-025-00574-5
PQID	3269074856
PQPubID	5642945
PageCount	23
ParticipantIDs	doaj_primary_oai_doaj_org_article_a1e8845110324ee588f20f091b21a768 proquest_journals_3269074856 crossref_primary_10_1007_s44163_025_00574_5 springer_journals_10_1007_s44163_025_00574_5
PublicationCentury	2000
PublicationDate	2025-12-01
PublicationDateYYYYMMDD	2025-12-01
PublicationDate_xml	– month: 12 year: 2025 text: 2025-12-01 day: 01
PublicationDecade	2020
PublicationPlace	Cham
PublicationPlace_xml	– name: Cham – name: Istanbul
PublicationTitle	Discover Artificial Intelligence
PublicationTitleAbbrev	Discov Artif Intell
PublicationYear	2025
Publisher	Springer International Publishing Springer Nature B.V Springer
Publisher_xml	– name: Springer International Publishing – name: Springer Nature B.V – name: Springer
References	T Fawcett (574_CR12) 2006; 27 574_CR18 574_CR16 C Bekkaye (574_CR3) 2025; 14 J Ho (574_CR21) 2020; 33 KP Murphy (574_CR34) 2012 A Sharma (574_CR44) 2024; 6 574_CR19 P Karadayı Ataş (574_CR23) 2024; 12 574_CR25 574_CR24 F Aslam (574_CR1) 2022; 62 574_CR22 F Liu (574_CR30) 2022; 15 J Singh (574_CR48) 2025; 22 N Dhieb (574_CR8) 2020; 8 DMW Powers (574_CR39) 2011; 2 M Hanafy (574_CR17) 2021; 99 A Kumar (574_CR28) 2025; 20 574_CR53 574_CR5 574_CR4 L Maiano (574_CR32) 2023; 231 SZS Nordin (574_CR36) 2024; 14 574_CR7 J Wei (574_CR55) 2021; 2021 574_CR56 P Patel (574_CR37) 2024; 24 PK Ataş (574_CR2) 2023; 24 I Kose (574_CR27) 2015; 36 A Kumar (574_CR29) 2024; 6 A Fernández (574_CR13) 2018 574_CR38 S Viaene (574_CR51) 2002; 69 574_CR43 574_CR41 574_CR40 574_CR47 574_CR46 574_CR45 A Gepp (574_CR14) 2012; 10 VA Fajardo (574_CR11) 2021; 169 CP Robert (574_CR42) 2004 B Efron (574_CR9) 1993 NV Chawla (574_CR6) 2002; 16 H He (574_CR20) 1997; 13 S Viaene (574_CR52) 2004; 16 Y Wang (574_CR54) 2018; 105 574_CR26 R Ming (574_CR33) 2024; 10 G Eom (574_CR10) 2023; 11 C Gomes (574_CR15) 2021; 88 574_CR31 M Sokolova (574_CR49) 2009; 45 574_CR35 S Viaene (574_CR50) 2005; 29
References_xml	– ident: 574_CR5 doi: 10.1175/1520-0493(1950)078<0001:VOFEIT>2.0.CO;2 – ident: 574_CR26 – ident: 574_CR38 doi: 10.1007/978-3-030-66665-1_17 – ident: 574_CR31 doi: 10.1109/ICDM.2008.17 – ident: 574_CR4 doi: 10.1023/A:1010933404324 – volume: 36 start-page: 283 year: 2015 ident: 574_CR27 publication-title: Appl Soft Comput doi: 10.1016/j.asoc.2015.07.018 – volume: 105 start-page: 87 year: 2018 ident: 574_CR54 publication-title: Decis Support Syst doi: 10.1016/j.dss.2017.11.001 – volume: 10 start-page: 537 year: 2012 ident: 574_CR14 publication-title: J Data Sci doi: 10.6339/JDS.201207_10(3).0010 – ident: 574_CR7 doi: 10.1145/2939672.2939785 – ident: 574_CR41 – volume: 2021 start-page: 6033860 year: 2021 ident: 574_CR55 publication-title: J Healthc Eng doi: 10.1155/2021/6033860 – volume: 10 year: 2024 ident: 574_CR33 publication-title: PeerJ Comput Sci doi: 10.7717/peerj-cs.2088 – volume: 2 start-page: 37 year: 2011 ident: 574_CR39 publication-title: J Mach Learn Technol – volume: 24 start-page: 333 year: 2024 ident: 574_CR37 publication-title: Int J Fuzzy Logic Intell Syst doi: 10.5391/IJFIS.2024.24.4.333 – volume: 6 year: 2024 ident: 574_CR44 publication-title: Eng Res Express doi: 10.1088/2631-8695/ad5d51 – ident: 574_CR45 doi: 10.1007/s11831-025-10271-2 – volume: 8 start-page: 58546 year: 2020 ident: 574_CR8 publication-title: IEEE Access doi: 10.1109/ACCESS.2020.2983300 – volume-title: An Introduction to the Bootstrap year: 1993 ident: 574_CR9 doi: 10.1007/978-1-4899-4541-9 – volume-title: Learning from imbalanced data sets year: 2018 ident: 574_CR13 doi: 10.1007/978-3-319-98074-4 – ident: 574_CR40 – ident: 574_CR16 – volume: 33 start-page: 6840 year: 2020 ident: 574_CR21 publication-title: Adv Neural Inf Process Syst – ident: 574_CR43 doi: 10.1007/978-3-030-96302-6_5 – ident: 574_CR56 doi: 10.1109/TII.2024.3366991 – ident: 574_CR22 doi: 10.14569/IJACSA.2020.0111054 – volume-title: Monte Carlo statistical methods year: 2004 ident: 574_CR42 doi: 10.1007/978-1-4757-4145-2 – volume: 99 start-page: 2819 year: 2021 ident: 574_CR17 publication-title: J Theor Appl Inf Technol – volume: 15 start-page: 139 year: 2022 ident: 574_CR30 publication-title: Algorithms doi: 10.3390/a15050139 – ident: 574_CR24 – ident: 574_CR47 – ident: 574_CR19 doi: 10.1109/IJCNN.2008.4633969 – volume: 88 start-page: 591 year: 2021 ident: 574_CR15 publication-title: J Risk Insur doi: 10.1111/jori.12359 – ident: 574_CR53 doi: 10.1007/s10115-025-02377-7 – ident: 574_CR25 doi: 10.1109/ACCESS.2024.3468993 – volume: 20 year: 2025 ident: 574_CR28 publication-title: PLoS ONE doi: 10.1371/journal.pone.0325900 – volume: 45 start-page: 427 year: 2009 ident: 574_CR49 publication-title: Inf Process Manage doi: 10.1016/j.ipm.2009.03.002 – volume: 14 start-page: 1440 year: 2025 ident: 574_CR3 publication-title: Stat Optim Inf Comput doi: 10.19139/soic-2310-5070-2655 – volume: 16 start-page: 321 year: 2002 ident: 574_CR6 publication-title: J Artif Intell Res doi: 10.1613/jair.953 – ident: 574_CR46 doi: 10.1609/aaai.v30i1.9825 – volume: 62 year: 2022 ident: 574_CR1 publication-title: Res Int Bus Financ doi: 10.1016/j.ribaf.2022.101744 – volume-title: Machine learning: a probabilistic perspective year: 2012 ident: 574_CR34 – volume: 69 start-page: 373 year: 2002 ident: 574_CR51 publication-title: J Risk Insur doi: 10.1111/1539-6975.00023 – volume: 12 start-page: 295 year: 2024 ident: 574_CR23 publication-title: Mathematics doi: 10.3390/math12020295 – volume: 16 start-page: 612 year: 2004 ident: 574_CR52 publication-title: IEEE Trans Knowl Data Eng doi: 10.1109/TKDE.2004.1277822 – volume: 14 start-page: 911 year: 2024 ident: 574_CR36 publication-title: Int J Electr Comput Eng (IJECE) doi: 10.11591/ijece.v14i1.pp911-921 – volume: 29 start-page: 653 year: 2005 ident: 574_CR50 publication-title: Expert Syst Appl doi: 10.1016/j.eswa.2005.04.030 – volume: 11 start-page: 3605 year: 2023 ident: 574_CR10 publication-title: Mathematics doi: 10.3390/math11163605 – volume: 27 start-page: 861 year: 2006 ident: 574_CR12 publication-title: Pattern Recogn Lett doi: 10.1016/j.patrec.2005.10.010 – volume: 169 year: 2021 ident: 574_CR11 publication-title: Expert Syst Appl doi: 10.1016/j.eswa.2020.114463 – volume: 22 start-page: 195 year: 2025 ident: 574_CR48 publication-title: Quant InfraRed Thermogr J doi: 10.1080/17686733.2024.2349976 – volume: 6 year: 2024 ident: 574_CR29 publication-title: Eng Res Express doi: 10.1088/2631-8695/ad76f9 – volume: 231 year: 2023 ident: 574_CR32 publication-title: Expert Syst Appl doi: 10.1016/j.eswa.2023.120644 – volume: 24 start-page: 319 year: 2023 ident: 574_CR2 publication-title: BMC Bioinform doi: 10.1186/s12859-023-05443-5 – ident: 574_CR18 – ident: 574_CR35 – volume: 13 start-page: 329 year: 1997 ident: 574_CR20 publication-title: Expert Syst Appl doi: 10.1016/S0957-4174(97)00045-6
SSID	ssj0002857390
Score	2.310538
Snippet	Fraud claim detection in auto insurance remains a vital yet complex challenge, mainly due to imbalanced data sets, non-linear feature interactions, and the... Abstract Fraud claim detection in auto insurance remains a vital yet complex challenge, mainly due to imbalanced data sets, non-linear feature interactions,...
SourceID	doaj proquest crossref springer
SourceType	Open Website Aggregation Database Index Database Publisher
StartPage	313
SubjectTerms	Accuracy Algorithms Artificial Intelligence Auto insurance Automobile insurance Classification Comparative analysis Computer Science Datasets Engineering Feature selection Fraud detection Fraud prevention Generative hybrid models Health insurance Insurance claims Insurance fraud Isolation forest Machine leaning algorithms Medical research Methods Neural networks Oversampling methods Support vector machines Survival analysis
SummonAdditionalLinks	– databaseName: DOAJ Directory of Open Access Journals dbid: DOA link: http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV07TxwxELYQoqBJgARxyQVNQZdbsbdnr-3yiHhUJ4okorP8VI5iD90jEh0_PTPeXV4SoknrXY0sz9jzzdjzDWMn6DNslTAsEYilC855RZS3urBKBF2m2slJ22xCzmbq5kZfP2v1RW_CWnrgduFO7TgqRSRaJbr-GIVSKDmhl3PV2CJWptMXhT8Lpm5zykhIjOa7KplcK8cJehTUvZUKMHkhXniiTNj_AmW-uhjN_uZij33ogCJM2wnus63YHLCPfRMG6PbkJ_bQEkfTqQV_7qn-CnJ3mxUgHIW0tJsAIa7zi6sG5g3YzXoB9ASdWmpEoEQsWPBPLOBgO6ISWCT4PT0fweV0NsJRFDRPaUP5Nei5yOPqM_t1cf7zx1XRtVUoPN05FjZNUimTjq6uKus113WSTlZprK1yXAgbVNCBcz9GM5O1tLwqo5yEMrlSyTg5ZNvNoolHDBDs8LrW1pXach288wpDSOV5cJ5kDtj3fonNXcueYR55krNCDCrEZIUYMWBnpIXHP4n5Og-gPZjOHsx79jBgw16HptuOK4MYlZIAStQDNur1-vT57Sl9-R9T-sp2K7K7_AhmyLbXy038xnb83_V8tTzOhvsPghztmw priority: 102 providerName: Directory of Open Access Journals – databaseName: ProQuest Central dbid: BENPR link: http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV1NbxMxELWg5cCl5VOkFDQHbmTFxrHX9gmlKIVTVCFAvVn-hFx2Szap1Ft_ej2ON1GR4MLVu7Iszdh-M555j5B36c4wNKawhCcsXTHGKFLeqspI7lUdGyumW7EJsVjIy0t1URJufSmrHM7EfFD7zmGO_EOCGRjHSd58vPpdoWoUvq4WCY2H5BCZypKfH57NFxdfd1kWKrlIUX3plsk9cwwhSIUqrtiIySp-70bKxP330OYfD6T53jk__t8VPyFHBXHCbOsiT8mD0D4jx4OaA5TN_Zzcbhmo8fiDXzfYyAVZJqeHhGshrszGgw_rXLrVwrIFs1l3gLXsqM0RADO6YMDt6cTBFMYT6CL8mM3H8Hm2GKfRNNEyxg0m6mAgNQ_9C_L9fP7t05eq6DNUDh8vKxOnsRZRBdtQapxiqonCChonykjLODdeeuUZc5Pkr6IRhtE6iKmvo62lCNOX5KDt2vCKQEJNrGmUsbUyTHlnnUyxqHTMW4dzjsj7wUb6akvDoXeEy9miOllUZ4tqPiJnaMbdn0ihnQe61U9ddqQ2kyAlsrPVCVOGwKVMLhsTfLJ0YlIQNiKng1l12de93tt0RMaDY-w__31JJ_-e7TV5TNElc53MKTlYrzbhDXnkrtfLfvW2ePUdatr-hA priority: 102 providerName: ProQuest
Title	Generative hybrid models for fraud detection in auto insurance with a comparative analysis of VAE, GAN, and diffusion approaches
URI	https://link.springer.com/article/10.1007/s44163-025-00574-5 https://www.proquest.com/docview/3269074856 https://doaj.org/article/a1e8845110324ee588f20f091b21a768
Volume	5
hasFullText	1
inHoldings	1
isFullTextHit
isPrint
journalDatabaseRights	– providerCode: PRVAON databaseName: DOAJ Directory of Open Access Journals customDbUrl: eissn: 2731-0809 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0002857390 issn: 2731-0809 databaseCode: DOA dateStart: 20210101 isFulltext: true titleUrlDefault: https://www.doaj.org/ providerName: Directory of Open Access Journals – providerCode: PRVHPJ databaseName: ROAD: Directory of Open Access Scholarly Resources customDbUrl: eissn: 2731-0809 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0002857390 issn: 2731-0809 databaseCode: M~E dateStart: 20210101 isFulltext: true titleUrlDefault: https://road.issn.org providerName: ISSN International Centre – providerCode: PRVPQU databaseName: ProQuest Central customDbUrl: eissn: 2731-0809 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0002857390 issn: 2731-0809 databaseCode: BENPR dateStart: 20211201 isFulltext: true titleUrlDefault: https://www.proquest.com/central providerName: ProQuest – providerCode: PRVPQU databaseName: Publicly Available Content Database customDbUrl: eissn: 2731-0809 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0002857390 issn: 2731-0809 databaseCode: PIMPY dateStart: 20211201 isFulltext: true titleUrlDefault: http://search.proquest.com/publiccontent providerName: ProQuest – providerCode: PRVPQU databaseName: Research Library customDbUrl: eissn: 2731-0809 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0002857390 issn: 2731-0809 databaseCode: M2O dateStart: 20211201 isFulltext: true titleUrlDefault: https://search.proquest.com/pqrl providerName: ProQuest – providerCode: PRVAVX databaseName: SpringerOpen customDbUrl: eissn: 2731-0809 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0002857390 issn: 2731-0809 databaseCode: C24 dateStart: 20211201 isFulltext: true titleUrlDefault: https://link.springer.com/search?facet-content-type=%22Journal%22 providerName: Springer Nature
link	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnV1NbxMxEB1BywEOLRQQaUs0B25kpd2NvbaPaZUCh4YIASony58QDhuUTZC48dNrO96UIjjAZQ-217Lssf1mPPMG4EW4M1Ttg1pCA5YuCCF1pLwVheLUitI3mo23ySbYbMavrsQ8B4V1vbd7_ySZTupdsBuJ2KGI6VdjBCUp6F3YpxUX0ZHvPMc4fE3mIsqCJp8jZP78661bKJH130KYvz2Kprvm4vD_RvkQDjK2xMlWGB7BHdcewWGftwHzNj6CB7-QED6Gn1vm6Xjs4ZcfMYALU3qcDgOeRb9SG4vWrZPLVouLFtVmvcTowx5zcjiMllxUaG5oxFFlphNcevw4mY7w1WQ2CqWho4X3m2igw57M3HVP4MPF9P356yLnZShMfLQslB_7knnhdFPXyggiGs80q30lFNeEUmW5FZYQUwU5ZQ1TpC4dG9vS65IzN34Ke-2ydc8AA1oiTSOULoUiwhpteNBBuSFWm9jnAF726yS_bek35I5oOU21DFMt01RLOoCzuJS7lpE6OxUsV59l3olSVY7zyMpWBizpHOU8iKoPsEnXlQrK1wBOe0GQeT93MoDcaEXgtBnAqF_4m-q_D-n435qfwP06yk7ylzmFvfVq457DPfN9vehWQ9g_m87m74ZJ3ofJfBC-l_XbUDN_czn_dA3fB_w5
linkProvider	Springer Nature
linkToHtml	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMw1V3LbtQwFL0qUyTYUJ5ioIAXsGIiMo4T2wuEBmjpqO1oFgWVlev4AbNJyjxA3fFFfCO-TjKjIsGuC7ZOZEXO8fW5tu85AM_DmqGpD2lJHrh0whijKHkrEy1yK1NflDxrzCb4ZCJOT-V0C351tTB4rbKLiTFQ29rgHvmrQDMwjxN58eb8W4KuUXi62lloNLA4dBc_Qsq2eD1-H_7vC0r3907eHSStq0Bi8Mgt0T7zKffSlQWl2kgmC89LTv1QalGyPNdWWGkZM8MwyrzgmtHU8cymvkwFd1no9xpsMwR7D7an4-Pp5_WuDhU5z2TaVufEGj2GlCdB11gs_GRJfmkFjEYBl9jtHweycZ3b3_nfRug23GoZNRk1U-AObLnqLux0bhWkDV734GejsI3hnXy9wEI1Em2AFiTwduLnemWJdct4Na0is4ro1bImeFcfvUccwR1roonZyKUT3Sq6kNqTT6O9AfkwmgxCa-ho5v0KNyJJJ9ruFvfh45WMwwPoVXXlHgIJrJAVhdRlKjWT1pRGhFxbGGZLg3324WWHCXXeyIyotaB0RJAKCFIRQSrvw1uEzfpNlAiPDfX8i2ojjtJDJwSqz6WBMzuXCxGmpA_0sKRDHZLMPux2MFJt3FqoDYb6MOiAuHn890969O_ensGNg5PjI3U0nhw-hpsUp0O8E7QLveV85Z7AdfN9OVvMn7YzisDZVUP0NyU1XIs
linkToPdf	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMw1V1Nj9MwEB0tXYS4sHyKwgI-wIlGmzpObB8QKuwWqoWqB0DLyev4A3pJln6A9sbv4tfhcZJWiwS3PXBNI6tynsdv7Jn3AJ6GPUNTH9KSPHDphDFGUfJWJlrkVqa-KHnWmE3w6VScnMjZDvzqemGwrLKLiTFQ29rgGflBoBmYx4m8OPBtWcTscPzy7FuCDlJ409rZaTQQOXbnP0L6tnwxOQzf-hml46MPr98mrcNAYvD6LdE-8yn30pUFpdpIJgvPS079UGpRsjzXVlhpGTPDMOO84JrR1PHMpr5MBXdZGPcK7AZKzmgPdmeT97PPmxMeKnKeybTt1In9egzpT4IOstgEypL8wm4YTQMuMN0_Lmfjnjfe-59n6ybcaJk2GTVL4xbsuOo27HUuFqQNanfgZ6O8jWGffD3HBjYS7YGWJPB54hd6bYl1q1iyVpF5RfR6VROs4UdPEkfwJJtoYrYy6kS3Si-k9uTT6GhA3oymg_A0DDT3fo0HlKQTc3fLu_DxUubhHvSqunL3gQS2yIpC6jKVmklrSiNCDi4Ms6XBMfvwvMOHOmvkR9RGaDqiSQU0qYgmlffhFUJo8yZKh8cH9eKLaiOR0kMnBKrSpYFLO5cLEZaqD7SxpEMdks8-7HeQUm08W6otnvow6EC5_fnvf-nBv0d7AtcCLtW7yfT4IVynuDJiqdA-9FaLtXsEV8331Xy5eNwuLgKnl43Q3z51ZUs
openUrl	ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Generative+hybrid+models+for+fraud+detection+in+auto+insurance+with+a+comparative+analysis+of+VAE%2C+GAN%2C+and+diffusion+approaches&rft.jtitle=Discover+Artificial+Intelligence&rft.au=Bekkaye%2C+Chadia&rft.au=Oukhouya%2C+Hassan&rft.au=Zari%2C+Tarek&rft.au=Guerbaz%2C+Raby&rft.date=2025-12-01&rft.pub=Springer+International+Publishing&rft.eissn=2731-0809&rft.volume=5&rft.issue=1&rft_id=info:doi/10.1007%2Fs44163-025-00574-5&rft.externalDocID=10_1007_s44163_025_00574_5
thumbnail_l	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2731-0809&client=summon
thumbnail_m	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2731-0809&client=summon
thumbnail_s	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2731-0809&client=summon