AI-Driven Detection Mechanism for UPI Fraud and QR Code Tampering

Fraud detection has become increasingly difficult due to the growing number of UPI transactions, necessitating sophisticated security measures. This work uses state-of-the-art Machine Learning Techniques to detect fraudulent transactions and detect tampering with QR codes in an AI-driven manner. The Random Forest algorithm's ability to differentiate between authentic and questionable transactions is assessed for UPI fraud detection. To find abnormalities, this model examines transaction behavior. Methods such as Random Forest, XGBoost, and Isolation Forest are used to detect unauthorized changes and irregularities in order to detect QR code tampering. These algorithms increase security and detection accuracy by continuously learning from transaction data and changing fraud tendencies. The goal of this research is to develop a strong fraud detection system driven by AI that improves transaction security and lowers financial risks for users