FRAUD DETECTION IN BANK PAYMENT SYSTEMS USING MACHINE LEARNING

Abstract

Fraudulent activities in banking systems have increased with the rapid growth of digital payments, leading to significant financial losses and reduced customer trust. Detecting these fraudulent transactions quickly and accurately has become essential for banks to maintain secure payment environments. Traditional manual methods are slow and ineffective; creating a strong need for automated and intelligent fraud detection systems this project focuses on developing a machine learning–based model to identify fraudulent bank payments. Transaction data is analyzed using important features such as amount, time, transaction type, and customer behavior patterns. Various algorithms, including Logistic Regression, Decision Trees, and Random Forest, are implemented and compared to determine which model provides the best accuracy and reliability for fraud detection. The final system classifies transactions as legitimate or fraudulent with improved precision, allowing banks to respond faster and prevent losses. By reducing manual effort and enabling real-time detection, the machine learning approach supports safer digital banking and strengthens overall financial security.