Abstract :

Online payment systems have become an essential part of digital transactions, but they are increasingly vulnerable to fraudulent activities. This project focuses on detecting online payment fraud using data-driven and machine learning techniques. By analyzing transaction patterns such as amount, location, time, and user behavior, fraudulent transactions can be identified effectively. The system aims to reduce financial loss and enhance transaction security. Supervised learning models are trained on historical transaction data to distinguish between genuine and fraudulent payments. The proposed approach improves detection accuracy while minimizing false alerts. This system is suitable for real-time fraud monitoring in digital payment platforms.