Comparative Study of Machine Learning Algorithms for Credit Card Fraud Detection

Abstract

Abstract: In the present world, advancement in technologies like e-commerce and financial technology (FinTech) has led to a surge in the daily volume of online card transactions and there are a lot of issues with credit cards. Identifying CC scams is currently among the most typical problems worldwide. An individual's credit card information may be fraudulently obtained and used by criminals for fraudulent purposes. Due to this significant increase in credit card fraud that impacts banks, merchants, and card issuers, it is essential to develop mechanisms that ensure the security and integrity of credit card transactions. This research examines divergent approaches to detecting credit card fraud using machine learning (ML) Models. The algorithms used are the Random Forest algorithm, Decision Trees, and the AdaBoost algorithm. The outcomes of these algorithms are based on accuracy, precision, recall, and F1-score and AUC-ROC score. Different models are compared and the algorithm that has good evaluation metrics is considered the best algorithm that is used to detect fraud