Abstract

Abstract: In this paper we mainlyfocus on credit card fraud detection in real world. Here the credit card frauddetection is based on fraudulent transactions. Generally credit card fraudactivities can happen in both online and offline. But in today's world onlinefraud transaction activities are increasing day by day. So in order to find theonline fraud transactions various methods have been used in existing system. Inproposed system we use Random Forest Algorithm(RFA) for finding the fraudulenttransactions and the accuracy of those transactions. This algorithm is based onsupervised learning algorithm where it uses decision trees for classificationof the dataset. After classification of dataset a confusion matrix is obtained.The performance of Random Forest Algorithm is evaluated based on the confusionmatrix. The results obtained from processing the dataset gives accuracy ofabout 90%.

Keywords:-CreditCard Fraud Detection, Transactions, Classification technique, Random ForestAlgorithm.