Abstract
Credit cards are very useful to make payments nowadays. Security is a major issue in making transactions using credit cards. Lots of frauds happen in different ways using credit cards. Machine learning algorithms are very useful in making predictions. In this paper, machine learning algorithms are used to analyze and predict the fraud in a credit card transaction. Different machine learning algorithms such as decision tree, random forest, linear regression, support vector machine, XG-Boost are applied on a standard data set. The prediction accuracy of these algorithms is measured and compared to each other. It is concluded that machine learning algorithms are very useful in predicting frauds using credit cards. The prediction accuracy of XG-Boost and the random forest is best among all.
