Credit Card Fraud Detection using Machine Learning: A Study

TL;DR

This paper reviews various machine learning techniques for credit card fraud detection, analyzing their effectiveness and comparing their advantages and disadvantages to improve financial security.

Contribution

It provides a comprehensive comparison of multiple machine learning methods for credit card fraud detection, highlighting their strengths and weaknesses.

Findings

Support Vector Machines perform well in fraud detection.

Random Forests offer high accuracy and robustness.

Trade-offs exist between model complexity and interpretability.

Abstract

As the world is rapidly moving towards digitization and money transactions are becoming cashless, the use of credit cards has rapidly increased. The fraud activities associated with it have also been increasing which leads to a huge loss to the financial institutions. Therefore, we need to analyze and detect the fraudulent transaction from the non-fraudulent ones. In this paper, we present a comprehensive review of various methods used to detect credit card fraud. These methodologies include Hidden Markov Model, Decision Trees, Logistic Regression, Support Vector Machines (SVM), Genetic algorithm, Neural Networks, Random Forests, Bayesian Belief Network. A comprehensive analysis of various techniques is presented. We conclude the paper with the pros and cons of the same as stated in the respective papers.