Securing Financial Transactions through Cutting-Edge Machine Learning Approaches to Effectively Combat Credit Card Fraud

The rise in credit card transactions has made fraud detection more important. While credit cards offer convenience, they also increase the risk of fraud, posing threats to financial security and consumer trust. To address this challenge, in this work, six machine learning (ML) algorithms were examined: Logistic Regression (LR), Random Forest (RF), K-Nearest Neighbours (KNN), Decision Tree (DT), Support Vector Machine (SVM), and Naive Bayes (NB). A dataset of 568,630 transactions from European cardholders in 2023 was utilized, featuring 30 attributes. Pre-processing techniques such as feature scaling and class balancing were applied to balance the data and improve accuracy. Detailed data analysis was performed to identify important patterns among the features. Each algorithm underwent training and optimization through hyperparameter tuning and cross-validation. Performance was evaluated using precision, accuracy, recall, F1 score, and Area under the ROC Curve (AUC). Results indicated that RF and LR performed best. High accuracy and AUC values were attained by RF, whereas LR successfully identified fraudulent and authentic transactions. These insights highlight the value of ML in enhancing fraud detection systems. As credit card use continues to grow, these tools will be essential for improving financial security.