A Survey on Detection of Fraudulent Credit Card Transactions Using Machine Learning Algorithms

Fraudulent transaction in credit cards has frequently rise in couple of years. Credit card fraud is a major issue for financial organizations, and accurate fraud detection is often difficult. Over Fifty percent of Americans have encountered a fraudulent transaction on their debit or credit card, and more than 1/3 of those who use these cards have done so repeatedly, according to 2021 yearly research. This translates to one hundred and twenty-seven million Americans who have at least once experienced credit card theft. Detection of such fraud happening over huge database is very difficult and time consuming using conventional method. By taking help of AI technology and developing an automated fraud detection system to detect and classify such mishappening using machine learning is an efficient way to deal with this kind of problem. This paper reviews various researchers work on detection of credit card frauds on highly imbalance dataset and discusses some machine learning techniques as Random Forest, Logistic Regression, SVM, Naive Bayes, XGBoost and KNN which are generally used by various researchers to build a model. The findings obtained from various researchers work showed that ensemble machine learning technique such as XGBoost and Random Forest are more capable of providing all over good performance in classifying such fraudulent and non-fraudulent transactions in credit cards.