Fraud Detection During Financial Transactions Using Machine Learning and Deep Learning Techniques

Abstract

Credit card fraud detection is a popular challenge in online transactions. Due to stolen cards that are used in unauthorized transactions by fraudsters, credit card fraud may occur. Also, a fraudster may utilize credit card credentials for his own purpose. As we are dealing with credit card issues, it is vital to build an effective credit card fraud detection system. Nowadays, deep learning and machine learning techniques are introduced in order to lessen human effort and time. In this paper, we present different experiments and findings related to the detection of credit card counterfeit detection. Here, Convolutional Neural Network(CNN), Convolutional Neural Network(CNN) with Gated Recurrent Units(GRU) and Adaptive Boosting (AdaBoost) algorithms are compared. We apply an oversampling technique, Synthetic Minority Oversampling Technique (SMOTE) to overcome the imbalance issue of our dataset. Most notably, Convolution Neural Network achieves very high AUC-ROC, accuracy, precision, and recall among all of them. In terms of correctness, this result is better than previously published work.