Machine Learning Credit Card Fraud Detection System

Abstract

The Credit Card Fraud Detection system is a web-based fraud detector tool that can be used to flag potential fraud cases in daily transactions. It is vital that credit card companies are able to spot fraudulent transactions using the credit card so that customers are not charged for items that they did not purchase. This article illustrates the modeling of a data set using past credit card transactions with the data of the ones that turned out to be fraudulent which is achieved using machine learning. The model recognizes whether a new transaction is fraudulent or not by validating a user before any transaction is being made through sending a One-Time-Password (OTP) to the user, hence detecting 90% to 100% of the fraudulent transactions thus minimizing financial crime. The Structured System and Design Methodology (SSADM) technique was employed in developing this system. In this process, we have focused on analyzing and pre-processing data sets as well as the deployment of multiple anomaly detection algorithms such as the Isolation Forest algorithm and Local Outlier Factor algorithm. The model was built with Python and implemented in an e-commerce site which was built with HTML, CSS, JavaScript, and its database as SQLite. After building and testing the system, it was discovered that in order for the system to be more accurate and precise, more transactional data need to be fed to the system.