International Journal of Scientific Research and Engineering Development

FinSecure: A Privacy-Preserving Federated Learning Framework for Financial Fraud Detection, The rapid digitalization of the financial sector has led to a significant rise in transaction volumes. At the same time, it has created new vulnerabilities for financial fraud. Traditional fraud detection systems rely on centralized machine learning models that need to pool sensitive customer data into a single server. This method raises important privacy concerns, data security risks, and regulatory compliance issues, such as GDPR and CCPA. This paper proposes FinSecure, a decentralized fraud detection framework based on Federated Learning (FL). Unlike centralized systems, FinSecure allows multiple financial institutions to collaboratively train a global fraud detection model without sharing raw transaction data. The system uses a Client-Server architecture where a central server collects model updates (gradients) from participating clients (banks) while keeping the actual data local and private. We implemented the system using FastAPI for the aggregation server, React for the dashboard, and TensorFlow for local model training. Our results show that FinSecure achieves accuracy similar to centralized models while ensuring complete data privacy.

Sajjan Udar, Nishant Ghuse,"FINSECURE: Federated Learning Platform for Fintech Companies" International Journal of Scientific Research and Engineering Development, V9(1): Page(810-816) Jan-Feb 2026. ISSN: 2581-7175. www.ijsred.com. Published by Scientific and Academic Research Publishing.