Enhancing Credit Card Fraud Detection: A Comparative Analysis of Machine Learning Models

A major issue of concern in the financial sector, credit card fraud puts risks to financial institutions as well as consumers. Machine learning methods have become rather successful instruments for spotting fraudulent transactions in recent years. This work compares the detection of credit card fraud employing two well-known machine learning techniques: Artificial Neural Networks (ANNs) and XGBoost. With a publicly available credit card transaction dataset, this work evaluates the accuracy, precision, recall, and F1 score of these algorithms. Furthermore taken into consideration to evaluate their fit for real-time fraud detection systems are the computational efficiency and scalability of ANNs and XGBoost. Among the five suggested methods, ANN shows best with an accuracy of 96.9%; among the classifiers, XGBoost shows best with an accuracy of 92.7%. The findings give financial institutions trying to install or enhance fraud detection systems direction as well as insights on the strengths and constraints of every method. This study supports the continuous initiatives to improve financial security by means of advanced machine learning approaches and fight credit card fraud.