E-Commerce Fraud Detection Based on Machine Learning

The rapid expansion of the e-commerce industry, particularly during the COVID-19 pandemic, has led to a significant rise in digital fraud and financial losses. Ensuring a secure e-commerce environment now demands effective cybersecurity and fraud prevention systems. However, research into fraud detection faces challenges due to limited availability of real-world datasets. Recent advancements in artificial intelligence (AI), machine learning (ML), and cloud computing have renewed interest in this area, but existing literature often lacks depth in evaluating ML algorithms specifically within e-commerce platforms like eBay and Facebook. Many reviews offer generalized insights but fail to capture how these techniques apply uniquely to digital marketplaces. To address this gap, our study adopts the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) methodology for a structured literature review. We aim to assess how ML and data mining techniques are applied to detect fraud in e-commerce. Analyzing 101 relevant publications from the past decade, our review identifies key trends, research opportunities, and the growing use of artificial neural networks in this field. These findings offer valuable insights to researchers and industry professionals seeking to implement effective fraud detection systems and highlight future research directions in combating e-commerce fraud.