Fraud detection in Online Payment Transaction using Machine Learning Algorithms

Abstract

Online payment transaction is a transaction in which payment is made using digitalized currency. Customers all over the world prefer online payments to purchase almost everything from furniture to clothing, from food to medicines, from gadgets to appliances, and whatnot. the online transaction has now evolved into many platforms. It is one of the most efficient methods provided by many companies to its customer. It not only helps build the company's revenue but also impacts the growth of the company. Let us not forget that with pros come the cons with the greatest advantage of having tedious transactions done at the fingertips comes the fear of fraudulent activity in which our hard-earned money can get theft within seconds. These activities can be determined by machine learning algorithms by feeding adequate data about these transactions. We have used machine learning algorithms such as SVM (Support Vector Machine), LR (Logistic regression), Naive Bayes, Decision tree, and Random Forrest for the same. It is observed that Random Forest classifier has outperformed comparing to other classifiers with the accuracy of 99.94%