DETECTING FRAUD IN UPI TRANSACTIONS: A STUDY USING RANDOM FOREST AND XGBOOST

DETECTING FRAUD IN UPI TRANSACTIONS: A STUDY USING RANDOM FOREST AND XGBOOST

*Geetangali  Bansod,  Shivani  Kardule, Pooja Gorade, Firdous Attar, Arpita  Wadavane

Dept. of Computer Engineering, KJ College of Engineering and Management Research, Pune, India

*Corresponding Author Mail ID: geethanjalibansod.kjcoemr@kjei.edu.in

Abstract
The widespread adoption of the Unified Payments Interface (UPI) has simplified digital transactions, enabling fast and easy payments. However, with its increasing usage comes a surge in fraudulent activities. This project presents a cutting-edge method for detecting UPI fraud by leveraging machine learning models and analyzing user behavior patterns. Utilizing a rich dataset that combines transaction records, behavioral data, and prior fraud cases, we develop a highly effective fraud detection system. The model uses techniques such as anomaly detection and pattern recognition to bolster the security of UPI transactions. Our experimental findings show a substantial enhancement in detecting fraudulent activities when compared to traditional approaches, underscoring the system’s potential to mitigate financial losses and build trust in digital payment systems.

Keywords: UPI Transactions, XG boost, Random Forest, Fraud Detection.