UPI Fraud Detection Using Machine Learning

Authors: Vijay Bhaskar, Abhishek, Hritik, Manjunath, Mrs. Sukanya

Abstract: The rapid increase in digital transactions, particularly through the Unified Payments Interface (UPI), has been accompanied by an alarming rise in fraudulent activities targeting unsuspecting users. This project aims to address this gap by developing an effective fraud detection system using machine learning. The proposed system utilizes four popular machine learning algorithms: Random Forest, Logistic Regression, Decision Tree, and Support Vector Machine (SVM) to classify transactions as either legitimate or fraudulent. By analyzing historical transaction data, the system detects anomalies or suspicious behavior indicative of fraud and flags such transactions in real-time. Experimental results indicate that machine learning offers the advantage of continuously improving detection accuracy as the system learns from new transaction data.