Toward smart financial security: design of web-based credit card fraud detection system

Abstract

As financial transactions become increasingly digital, there is an urgent need for stronger security measures. This paper presents a web-based credit card fraud detection system designed to address this challenge and enhance transaction security. The system was designed to provide an easy tool for identifying suspicious activity without relying on complex predictive models such as machine learning. It uses a Python backend linked to a SQLite database and a browser-based interface built with HTML, CSS, and JavaScript. The web-based credit card fraud detection system employs rule-based validation and anomaly checks that examine transaction amounts, frequency of use, and deviations from a cardholder’s normal behavior. Evaluation involved unit, integration, performance, and security testing. The results suggest that the system performs well by identifying questionable transactions, maintaining stable operation under high transaction volumes, and resisted common security threats. While its rule-based design may limit adaptability to evolving fraud tactics, the system demonstrates that a straightforward web application can still offer meaningful protection for digital financial operations.