ANALYSIS AND COMPARISON OF VARIOUS MACHINES LEARNING ALGORITHM FOR CREDIT CARD FRAUD DETECTION

Abstract

Online shopping has become an integral part of our life. As card payment becomes the most prevailing mode of payment for both online as well as regular purchase, frauds related to it are also accelerating. Fraud detection in card payment is a crucial part of e-shopping. It includes monitoring of the spending behavior of customers in order to detect and avoid fraud. This pa-

per uses Machine Learning Models which are built with various algorithms. It uses classification algorithms such as Decision Tree and Random Forest for better efficiency. Deep learning neural networks use feature extraction and Back propagation to improve accuracy of results from the training data. Random Forest is an ensemble learning approach which contains a combination of many algorithms. Random Forest uses training data to come up with a solution, it generates multiple decision trees and selects the decision trees which get to the best solution. The goal of this paper is to check the performance of Decision Tree, Random Forest and Deep Learning algorithms on highly skewed credit card fraud details. Based on precision, sensitivity, specificity and accuracy, the efficiency of the techniques is assessed.