Privacy-Preserving Federated Fraud Detection System for Collaborative Banking

·                 This paper presented a federated learning system that lets multiple banks work together to strengthen a shared fraud detection model while each bank keeps exclusive control of its own transaction data. By limiting all communication between banks to model parameters, and combining those parameters through Federated Averaging at a central server, the system lets fraud knowledge learned at one bank help detection at another bank, something a conventional, siloed fraud detection setup cannot do. ·                 Built across three simulated bank nodes using PyTorch, Flower, and a Streamlit interface, the system trains reliably at the local level and produces a global model that genuinely reflects knowledge pooled from all three banks. The results support the central claim of this project: banks do not have to choose between protecting customer privacy and detecting fraud effectively. Extending the system with differential privacy, secure aggregation, and explainability, as outlined in Section VII, would move it closer to something a real bank could deploy.

[1]                    Y. Tang and Z. Liu, "A Credit Card Fraud Detection Algorithm Based on SDT and Federated Learning," IEEE Access, 2024. [2]                    P. Gawade, J. Hande, and P. Kene, "Federated Learning in Banking: A Secure and Intelligent Approach to Financial Fraud Detection," Journal of Emerging Technologies and Innovative Research (JETIR), June 2025. [3]                    A. A. Ahmed and O. O. Alabi, "Secure and Scalable Blockchain-Based Federated Learning for Cryptocurrency Fraud Detection: A Systematic Review," IEEE Access, July 2024. [4]                    T. El Hallal and Y. El Mourabit, "Federated Learning for Credit Card Fraud Detection: Key Fundamentals and Emerging Trends," IEEE, June 2024. [5]                    S. K. Aljunaid, S. J. Almheiri, H. Dawood, and M. A. Khan, "Secure and Transparent Banking: Explainable AI-Driven Federated Learning Model for Financial Fraud Detection," Journal of Risk and Financial Management, March 2025. [6]                    W. Zheng, L. Yan, C. Gou, and F.-Y. Wang, "Federated Meta-Learning for Fraudulent Credit Card Detection," IJCAI, 2019. [7]                    C. H. Kennedy, A. Hilal, and M. Momeni, "The Role of Federated Learning in Improving Financial Security: A Survey," IEEE, October 2025. [8]                    N. Abid, "Improving Accuracy and Efficiency of Online Payment Fraud Detection and Prevention with Machine Learning Models," International Journal of Innovative Science and Research Technology, December 2025. [9]                    A. R. Gudimetla, "Decentralized Fraud Detection: Advancing Payment Security Through Federated Learning," IJITMIS, January 2025. [10]                 M. S. Islam and N. Rahman, "AI-Driven Fraud Detections in Financial Institutions: A Comprehensive Study," Journal of Computer Science and Technology Studies, January 2025.