Abstract

The increasing volume of online transactions has heightened the risk of fraud, making real-time fraud detection crucial for safeguarding financial systems. This paper explores the development and application of machine learning (ML) models for detecting fraudulent activities in real-time online transactions. The study investigates various ML algorithms, including supervised and unsupervised learning techniques, to identify patterns indicative of fraud. We evaluate the performance of different models based on accuracy, precision, recall, and F1-score. The results show that ensemble methods and deep learning techniques outperform traditional approaches in terms of accuracy and detection speed. The study also emphasizes the importance of data preprocessing, feature selection, and real-time model deployment for achieving robust fraud detection systems. This research contributes to the ongoing efforts to enhance online transaction security and provides insights into implementing effective fraud detection mechanisms using machine learning