UPIGuard: A Fraud Detection System for UPI Transactions

Abstract

The increasing adoption of digital payment platforms, particularly the Unified Payments Interface (UPI), has significantly transformed the way financial transactions are conducted. While UPI provides a fast, convenient, and widely accessible payment method, it has also become a target for various fraudulent activities such as phishing attacks, unauthorized access, and deceptive transaction requests. Detecting such fraudulent activities in real time has become a critical challenge for financial systems. This research proposes a machine learning-based UPI fraud detection system designed to analyze transaction patterns and identify suspicious activities before they result in financial loss. The system evaluates multiple transaction attributes including transaction amount, time, user behavior, and device information to detect anomalies and predict the likelihood of fraud. By leveraging data-driven models and intelligent pattern recognition, the proposed system aims to improve the accuracy and efficiency of fraud detection while minimizing false alerts.