ENHANCED UPI FRAUD DETECTION

Abstract

With its smooth and quick financial transfers, the Unified Payments Interface's (UPI) explosive growth has transformed digital transactions. But as a result of this expansion, UPI fraud cases have increased, taking advantage of flaws in conventional fraud detection systems. Traditional fraud detection systems are unable to identify changing fraud patterns since they are based on threshold-based models and static rules. In order to analyze transaction behaviors and identify abnormalities in real time, this study proposes an enhanced UPI fraud detection system that makes use of machine learning techniques like decision trees, random forests, and neural networks. The suggested solution improves transaction security, lowers false positives, and increases the accuracy of fraud detection. It successfully detects and reduces fraudulent activity by combining adaptive learning and real-time monitoring, guaranteeing a safe digital payment ecosystem.