ABSTRACT

Credit card fraud detection is based on false transactions. Generally credit card fraud can be in both online and offline. Now online fraud activities are adding day by day. So in order to find the online fraud deals several algorithms have been used. In proposed system we utilize Random Forest Model (RFM) for predicting the fraudulent transactions and the exactness of those transactions. This algorithm is grounded on supervised learning algorithm where it uses decision trees for division of the dataset. After division of dataset a confusion matrix is attained. The performance of Random Forest Algorithm is assessed based on the confusion matrix. The results attained from recycling the dataset gives accuracy of above 90%.