Enhancing Credit Card Security: A Machine Learning Approach for Fraud Detection

Transactions increase in frequency as the world swiftly adopts digitisation and a cashless economy. The use of credit cards has dramatically increased in recent years. Financial institutions face a large loss as a result of the expanding related fraud activities. As a result, we must evaluate transactions to determine which are real and which are fraudulent. In this essay, we present a comprehensive study of the various methods for detecting credit card fraud. These methods make use of Bayesian Belief Networks, algorithms, Decision Trees, Logistic Regression, Genetic Random Forests, Markov Models, and Hidden Support Vector Machines (SVM). There is a thorough examination of several approaches offered. Following analysis, we put the SVM and LR models into practice, along with addressing data imbalances and feature extraction. We summarise the advantages and disadvantages of the same as stated in the respective publications as our conclusion to the paper. We want to use a credit card approach that is effective.