Counterfeit Detection in the e-Commerce Industry Using Machine Learning: A Review

Abstract

The past decade has experienced an exponential rise in online purchases along with using credit cards and associated financial tools. This widespread e-commerce use has resulted in an unprecedented surge in frauds that range from financial frauds to fake online-shop frauds. Consequently. The detection and safeguarding of users from such frauds have been a vital goal to achieve for many organizations and enterprises, most of which aim to achieve the same through the application of machine learning to build classifier models that detect and classify data (transactions, online shops, and other e-commerce data) into fraudulent and legit classes. This survey paper aims to understand the advancements made in the last decade in the field to understand the progress made along with the gaps associated with the current research work. Moreover, the hurdles or challenges pertaining to widespread implementation are also discussed with potential solutions and prospects comprehensively; while providing insights on the most feasible ML algorithm(s) based on the survey, followed by future directions of research work to make it equipped for real-world implementation.