Credit card application form identity crime detection using Data mining algorithm with multilayer algorithm

Abstract

Due to market uncertainties, slowing economic growth, and the rapid rise of online e-commerce, fraud has become a widespread issue. With the rapid advancement of electronic commerce technology, credit card usage has increased, making it the most popular payment method for both online and offline purchases. As a result, credit card fraud is on the rise, and customers who require smart cards and loans can now apply for credit cards online or by filling out paper applications. Unfortunately, these applications have uncovered instances of fraud, including identity theft, which has become a severe concern for both credit card customers and banks. Fraudsters are stealing customers' identities and obtaining credit cards, putting both customers and banks at significant risk.

However, existing business rules and scorecards-based non-data-mining tracing approaches and fraud recognition have been found to have flaws. To address these issues, this study proposes a real-time method for detecting fraudsters at the moment of application submission using a new multi-layer fraud tracing system based on data-mining algorithms. This system employs two algorithms: communal tracing and spike tracing, which work together to improve the accuracy, speed, and cost-effectiveness of fraud tracing. Before a credit card is issued, the application is validated at the time of submission to prevent fraudulent applications from being approved.