Fake Review prediction using Machine Learning in E-Commerce Platform

Abstract: The prevalence of fake reviews on e-commerce platforms poses a significant challenge, undermining consumer trust and distorting the marketplace. This study addresses the detection of fake reviews using machine learning techniques, focusing on both semi-supervised and supervised learning approaches. We implement the Expectation-Maximization (EM) algorithm alongside the Naive Bayes classifier to distinguish genuine reviews from fraudulent ones. Our system analyses review content and various features, including word frequency count, sentiment polarity, and review length, to enhance detection accuracy. By utilizing a combination of Visual Studio for the frontend and SQL Server for the backend, we develop a robust platform capable of real-time detection and reporting of fake reviews. The proposed solution aims to provide a more reliable and authentic review system, ultimately improving the consumer experience and integrity of e-commerce platforms.