ccfDetector: Utilizing GAN and Deep Learning for Credit Card Fraud Detection

In recent years, the widespread adoption of e-commerce has led to a corresponding growth in credit card fraud. To combat this issue, organizations have begun implementing fraud detection systems using machine learning techniques. However, most credit card fraud detection datasets suffer from imbalanced data, which can impact the performance of these systems. In this study, we propose using Generative Adversarial Networks (GANs) to address this problem. By training the GAN on the unbalanced dataset of credit card fraud detection, we can generate additional synthetic records belonging to the underrepresented class (e.g. fraudulent transactions) to balance the dataset. Moreover, we propose a deep learning model based on Artificial Neural Networks to classify the dataset. Also, we examine the impact of feature selection on the performance of the model and demonstrate that our approach results in improved sensitivity. We combined these three components in a system called ccffDetector. Our results show that using generated records merged with the training dataset results in the highest sensitivity of 0.9024%. This indicates that our approach has the potential to improve the credit card fraud detection system's effectiveness.