A Review: Identification of Credit Card Fraud Using Machine Learning and Anomaly Detection Approach’s on Imbalanced Data

Ayush Bilgaiyan; Vinod Patel

Abstract

Credit cards are among the most widely used payment methods in today's culture, and online purchases have become quite popular. Credit card fraud has emerged as a problem in this industry as a direct result of its popularity. The issue of credit card fraud is becoming a global problem. Credit card fraud has increased due to the widespread use of these payment methods. Thanks to credit card use, e-commerce has flourished, and the infrastructure for electronic payments has become more user-friendly. To combat fraud, machine learning techniques are being used on a larger scale.When it comes to inspecting customer data, ML algorithms are crucial. Security of credit card transactions and the efficiency of online banking are, therefore, primary concerns for financial organizations. They want to do this by creating more effective methods of detecting fraudulent transactions, which will lead to a reduction in overall fraud. This study aims to define fraud detection, provide fraud detection methods, discuss banking industry fraud difficulties and challenges, and outline contemporary solutions based on ML techniques. This work presents an analysis of the existing literature on machine learning and anomaly detection methods utilized in credit card fraud detection (CCFD).As far as ML is concerned, it is the best way to guarantee privacy while increasing CCFD accuracy.Different class imbalances, machine learning, data mining, as well as anomaly detection techniques have been reviewed and compared in the domain of fraud detection systems.This review provides the overview of CCFD, key features, trends, anomaly detection techniques, machine learning techniques, class imbalance problems and existing work on CCFD.

Abstract

Keywords