Machine Learning Models for Detecting Anomalies in Online Payment: A Comparative Analysis

With the growing use of online payment systems, the necessity for strong security measures to defend against fraudulent activity has become critical. Machine learning algorithms-based anomaly detection approaches have developed as efficient solutions for spotting aberrant patterns and detecting fraudulent transactions in online payment systems. It offers efficient and effective online payment monitoring, protecting against fraudulent activity. In the present study, the applications of machine learning techniques for anomaly detection in online payment system is investigated. In conclusion, results provided by four models namely, Logistic Regression, Decision Tree, Random Forest and Extreme Gradient Boosting (XGB) Classifier can be preferred for anomaly detection in online payment. Among the four models, the XGB Classifier provided the highest model accuracy.