An Explainable Machine Learning Framework for Real-Time Credit Card Fraud Detection using Random Forest and SHAP

IX. ConclusionThis paper presented an explainable, real-time credit card fraud detection framework that combines a Random Forest classifier with SHAP explanations and an operational web stack composed of FastAPI, Streamlit, PostgreSQL, and Redis. The system meets demanding non-functional targets — accuracy above 93% and sub-100 ms inference latency — while making each decision auditable through per-feature Shapley attributions. The integration of an alerting subsystem, three-strike account blocking, and a transaction simulator demonstrates that academic ML prototypes can be designed with production realism in mind. The framework can serve as a foundation for future research in concept drift handling, deep sequence models, graph-based fraud detection, and federated learning across financial institutions.

[1] S. M. Lundberg and S.-I. Lee, “A unified approach to interpreting model predictions,” in Proc. 31st Int. Conf. Neural Information Processing Systems (NIPS), 2017, pp. 4765–4774. [2] S. M. Lundberg et al., “From local explanations to global understanding with explainable AI for trees,” Nature Machine Intelligence, vol. 2, no. 1, pp. 56–67, 2020. [3] S. Bhattacharyya, S. Jha, K. Tharakunnel, and J. C. Westland, “Data mining for credit card fraud: A comparative study,” Decision Support Systems, vol. 50, no. 3, pp. 602–613, 2011. [4] L. Breiman, “Random forests,” Machine Learning, vol. 45, no. 1, pp. 5–32, 2001. [5] A. Dal Pozzolo, O. Caelen, R. A. Johnson, and G. Bontempi, “Calibrating probability with undersampling for unbalanced classification,” in Proc. IEEE Symp. Computational Intelligence and Data Mining (CIDM), 2015. [6] Reserve Bank of India, “Annual report on banking frauds in India,” 2023. [Online]. Available: https://www.rbi.org.in/ [7] S. Ramírez, “FastAPI — Modern, fast web framework for building APIs.” [Online]. Available: https://fastapi.tiangolo.com/ [8] Streamlit Inc., “Streamlit — The fastest way to build data applications.” [Online]. Available: https://streamlit.io/ [9] F. Pedregosa et al., “Scikit-learn: Machine learning in Python,” J. Mach. Learn. Res., vol. 12, pp. 2825–2830, 2011. [10] SHAP Contributors, “SHAP (SHapley Additive exPlanations) documentation.” [Online]. Available: https://shap.readthedocs.io/ [11] PostgreSQL Global Development Group, “PostgreSQL: The world’s most advanced open source database.” [Online]. Available: https://www.postgresql.org/ [12] Redis Ltd., “Redis — The real-time data platform.” [Online]. Available: https://redis.io/