Abstract
Credit card fraud is a significant concern for financial institutions and customers alike. To combat this issue, researchers have explored various techniques, including machine learning and deep learning, to develop effective fraud detection systems. This abstract provides an overview of the approach and key findings in Credit Card Fraud Detection using Machine Learning and Deep Learning. The study begins by collecting a comprehensive dataset containing transactional information, including features such as transaction amount, location, time, and customer demographics. This dataset incorporates both genuine and fraudulent transactions, enabling the development of a robust model capable of distinguishing between the two. Initially, traditional machine learning algorithms, such as logistic regression, decision trees, and random forests, are employed to build a baseline fraud detection system. These algorithms utilize various features and patterns extracted from the dataset to classify transactions as either genuine or fraudulent. The performance of these models is evaluated using metrics like accuracy, precision, recall, and F1 score. Subsequently, deep learning techniques, particularly deep neural networks, are applied to enhance the fraud detection system. Multiple layers of neurons are utilized to extract intricate patterns and relationships within the data. The network is trained using backpropagation and optimization algorithms, iteratively improving its ability to accurately classify transactions. The deep learning model's performance is evaluated and compared to the machine learning baseline. The results indicate that both machine learning and deep learning approaches are effective in detecting credit card fraud. However, deep learning models consistently outperform traditional machine learning algorithms, achieving higher accuracy, precision, recall, and F1 score. The deep neural network's ability to capture complex patterns and dependencies within the data contributes to its superior performance.