Performance Evaluation of Machine Learning Algorithms for Real-Time Credit Card Fraud Detection

In this research, a comparative analysis of multiple machine learning algorithms for credit card fraud detection was performed. The increasing use of digital payment systems has made fraud detection an important challenge in the financial sector. To address this issue, machine learning models including Logistic Regression, Decision Tree, Random Forest, and XGBoost were implemented and evaluated using a real-world credit card transaction dataset. The experimental results demonstrated that different algorithms show different performance levels in identifying fraudulent transactions. Among the evaluated models, ensemble learning algorithms such as Random Forest and XGBoost achieved better accuracy, precision, recall, and F1-score compared to traditional classification models. These algorithms successfully identified fraudulent patterns while minimizing false predictions. This study concludes that machine learning can significantly improve fraud detection systems by providing faster, more accurate, and intelligent decision-making. The findings of this research can help financial institutions build secure real-time fraud monitoring systems and reduce financial losses caused by fraudulent activities

[1] A. Dal Pozzolo, O. Caelen, R. A. Johnson, and G. Bontempi, “Calibrating Probability with Undersampling for Unbalanced Classification,” IEEE Symposium Series on Computational Intelligence, 2015. [2] J. West and M. Bhattacharya, “Intelligent Financial Fraud Detection: A Comprehensive Review,” Computers & Security, 2016. [3] S. Bhattacharyya, S. Jha, K. Tharakunnel, and J. C. Westland, “Data Mining for Credit Card Fraud Detection,” Decision Support Systems, 2011. [4] T. Chen and C. Guestrin, “XGBoost: A Scalable Tree Boosting System,” ACM SIGKDD, 2016. [5] L. Breiman, “Random Forests,” Machine Learning Journal, 2001. [6] Kaggle, “Credit Card Fraud Detection Dataset.”