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!
Popis
Shrnutí: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.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:2731-0809
2731-0809
DOI:10.1007/s44163-025-00574-5