A HYBRID ABNORMALITY–REGRESSION APPROACH FOR CREDIT CARD FRAUD PREDICTION IN BANKS USING WEB TECHNOLOGY

Abstract

Credit card fraud is one of the most critical challenges faced by the banking sector, leading to financial losses, reduced customer trust, and increased operational risks. Traditional fraud detection systems often rely on static rule-based mechanisms, which are inadequate in detecting evolving fraud patterns. This study proposes a hybrid fraud detection framework that integrates abnormality detection and regression algorithms within a webbased application to enhance fraud prediction in real time. The hybrid approach improves accuracy by combining anomaly recognition with statistical regression, thus identifying both unexpected patterns and behavioral deviations. The proposed system enables banks to monitor transactions, predict fraud likelihood, and provide alerts through an interactive web platform. Experimental results suggest that the hybrid system demonstrates higher accuracy, reduced false positives, and improved adaptability compared to conventional systems