Abstract
Fraudulent transactions are a persistent challenge for financial institutions, demanding robust detection systems to maintain customer trust. Key obstacles include the rarity of fraud cases, leading to imbalanced datasets, and strict privacy regulations limiting data sharing. Additionally, fraud detection must be transparent to preserve user trust. This research addresses these issues by combining Federated Learning (FL) and Explainable AI (XAI), allowing institutions to collaboratively train models without sharing data, thus protecting privacy while ensuring model transparency and interpretability.

This publication has 16 references indexed in Scilit: