Graph Neural Network-Based Detection of Organized Fraud in Insurance Claim Networks
Keywords:
Graph Neural Networks, Insurance Fraud Detection, Organized Fraud, Graph Convolutional Networks, Graph Attention Networks, Relational Learning, Financial Crime AnalyticsAbstract
Organized insurance fraud represents a significant financial and operational challenge for insurers worldwide, particularly when fraudulent activities are executed through coordinated networks of interconnected entities. Traditional machine learning models predominantly rely on independent claim-level features and therefore fail to capture the relational dependencies inherent in collaborative fraud schemes. This study proposes a Graph Neural Network (GNN)-based framework for detecting organized fraud within insurance claim networks by modeling entities and their interactions as a structured graph. The insurance ecosystem is represented as a heterogeneous graph in which nodes correspond to claimants, policies, medical providers, and repair facilities, while edges encode verified relational linkages. A hybrid architecture combining Graph Convolutional Networks (GCN) and Graph Attention Networks (GAT) is developed to jointly capture structural propagation and adaptive relational importance. The proposed model incorporates weighted binary cross-entropy to address severe class imbalance and leverages attention coefficients to enhance interpretability. Experimental evaluation demonstrates that the hybrid GAT–GCN model outperforms strong tabular baselines, including logistic regression, random forest, XGBoost, and multilayer perceptron classifiers. The proposed framework achieves superior AUC and F1-score performance, significantly improving fraud recall while maintaining controlled false positive rates. Ablation studies confirm the complementary contribution of convolutional and attention mechanisms, and statistical testing validates the robustness of performance improvements.The findings establish that organized insurance fraud is fundamentally a relational learning problem and that graph-based deep learning architectures provide a scalable, interpretable, and operationally viable solution for next-generation fraud detection systems.
Downloads
References
R. Y. Gupta, S. S. Mudigonda, and P. K. Baruah, "A comparative study of using various machine learning and deep learning-based fraud detection models for universal health coverage schemes," International Journal of Engineering Trends and Technology, vol. 69, no. 3, pp. 96-102, 2021.
A. Singh and A. Jain, "An efficient credit card fraud detection approach using cost‐sensitive weak learner with imbalanced dataset," Computational Intelligence, vol. 38, no. 6, pp. 2035-2055, 2022.
L. Wang, P. Li, K. Xiong, J. Zhao, and R. Lin, "Modeling heterogeneous graph network on fraud detection: A community-based framework with attention mechanism," in Proceedings of the 30th ACM international conference on information & knowledge management, 2021, pp. 1959-1968.
L. Akoglu, H. Tong, and D. Koutra, "Graph based anomaly detection and description: a survey," Data mining and knowledge discovery, vol. 29, no. 3, pp. 626-688, 2015.
T. N. Kipf and M. Welling, "Semi-supervised classification with graph convolutional networks," arXiv preprint arXiv:1609.02907, 2016.
P. Veličković, G. Cucurull, A. Casanova, A. Romero, P. Lio, and Y. Bengio, "Graph attention networks," arXiv preprint arXiv:1710.10903, 2017.
Y. Dou, Z. Liu, L. Sun, Y. Deng, H. Peng, and P. S. Yu, "Enhancing graph neural network-based fraud detectors against camouflaged fraudsters," in Proceedings of the 29th ACM international conference on information & knowledge management, 2020, pp. 315-324.
Y. Yoo, J. Shin, and S. Kyeong, "Medicare fraud detection using graph analysis: A comparative study of machine learning and graph neural networks," IEEE Access, vol. 11, pp. 88278-88294, 2023.
R. M. Zakaria, M. M. Rahman, H. Rahman, and M. A. Rafi, "Detecting Financial Fraud in Real-Time Transactions Using Graph Neural Networks and Anomaly Detection Techniques," Journal of Economics, Finance and Accounting Studies, vol. 7, no. 6, pp. 01-13, 2025.
Z. Ying, D. Bourgeois, J. You, M. Zitnik, and J. Leskovec, "Gnnexplainer: Generating explanations for graph neural networks," Advances in neural information processing systems, vol. 32, 2019.
Z. Qin, Y. Liu, Q. He, and X. Ao, "Explainable graph-based fraud detection via neural meta-graph search," in Proceedings of the 31st ACM international conference on information & knowledge management, 2022, pp. 4414-4418.
