Explainable Graph-Based AI for Real-Time Fraud Detection in Distributed Healthcare Claims Processing Systems
Abstract
Healthcare fraud has become a highly important problem, costing billions of dollars every year and making healthcare systems less trustworthy. As the claims become more and more digitized and the distributed processing architectures emerge, more complex and changing patterns of fraud cannot be detected using the more traditional rule-based fraud detection methods. The presented paper suggests a sophisticated AI-based model that combines graph-based learning, explainable artificial intelligence (XAI), and real-time event processing to identify fraudulent actions in healthcare claims in an effective and transparent way. The proposed system describes healthcare participants (patients, providers, insurers, and transactions) as graphs, where graph neural networks (GNNs) can identify hidden relationships and abnormal patterns. In order to be transparent and comply with regulatory requirements, explainability methods including SHAP (SHapley Additive exPlanations) and LIME (Local Interpretable Model-Agnostic Explanations) are implemented in the framework to enable the stakeholders to interpret the model decisions and verify fraud predictions. Moreover, the architecture will utilize real-time data streaming solutions such as Apache Kafka and AWS SQS to scale the processing of the claims events and allow detecting fraud and take action in real-time. The system is scaled to distributed settings, which guarantee scalability, resiliency, and smooth interoperability with the current healthcare infrastructure. Experimental analyses exhibit better detection, less false positives and higher interpretability than conventional machine learning methods. The suggested model can not only enhance the ability to detect fraud but also promote trust and responsibility by offering explainable knowledge and is appropriate in the context of modern, data-driven healthcare ecosystems.
References
1. Ngai, E. W. T., Hu, Y., Wong, Y. H., Chen, Y., & Sun, X. (2011). The application of data mining techniques in financial fraud detection: A classification framework and an academic review of literature. Decision Support Systems, 50(3), 559–569.
2. Bauder, R. A., & Khoshgoftaar, T. M. (2018). A survey of Medicare data processing and fraud detection techniques. Health Services and Outcomes Research Methodology, 18(1), 1–24.
3. Johnson, J. M., & Khoshgoftaar, T. M. (2019). Survey on deep learning with class imbalance. Journal of Big Data, 6(1), 1–54.
4. Esteva, A., Robicquet, A., Ramsundar, B., et al. (2019). A guide to deep learning in healthcare. Nature Medicine, 25(1), 24–29.
5. Lundberg, S. M., & Lee, S. I. (2017). A unified approach to interpreting model predictions. In Advances in Neural Information Processing Systems (NeurIPS) (pp. 4765–4774).
6. Ribeiro, M. T., Singh, S., & Guestrin, C. (2016). “Why should I trust you?” Explaining the predictions of any classifier. In Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 1135–1144).
7. Kipf, T. N., & Welling, M. (2017). Semi-supervised classification with graph convolutional networks. In International Conference on Learning Representations (ICLR).
8. Hamilton, W., Ying, Z., & Leskovec, J. (2017). Inductive representation learning on large graphs. In Advances in Neural Information Processing Systems (NeurIPS) (pp. 1024–1034).
9. Weber, M., Domeniconi, G., Chen, J., et al. (2019). Anti-money laundering in Bitcoin: Experimenting with graph convolutional networks for financial forensics. KDD Workshop on Anomaly Detection in Finance.
10. Chen, C., Li, C., & Luo, X. (2020). A real-time fraud detection framework based on streaming analytics. IEEE Access, 8, 123–135.
11. Kreps, J., Narkhede, N., & Rao, J. (2011). Kafka: A distributed messaging system for log processing. In Proceedings of the NetDB Workshop.
12. Amazon Web Services. (2023). Amazon SQS: Developer guide. Retrieved from https://docs.aws.amazon.com
13. Phua, C., Lee, V., Smith, K., & Gayler, R. (2010). A comprehensive survey of data mining-based fraud detection research. Artificial Intelligence Review, 34(1), 1–14.
14. Carcillo, F., Dal Pozzolo, A., Le Borgne, Y. A., et al. (2021). Combining unsupervised and supervised learning in credit card fraud detection. Information Sciences, 557, 317–331.
15. Zheng, Z., Xie, S., Dai, H., et al. (2018). Graph neural networks: A review of methods and applications. arXiv preprint arXiv:1812.08434.
