Modeling of Fake Product Review Authentication System

ABSTRACT Online reviews influence consumer decisions, shaping the trustworthiness of e-commerce platforms. However, the prevalence of fake reviews has become a critical issue, misleading customers and affecting business reputations. This paper presents an AI-driven Fake Product Review Monitoring and Removal System to identify and eliminate fraudulent reviews. By leveraging sentiment analysis, metadata tracking, and machine learning models, the system detects suspicious reviews with high accuracy. The methodology integrates natural language processing (NLP) techniques and behavioural analysis to enhance detection efficiency. The system’s real-time processing capabilities ensure scalability, making it adaptable for large-scale platforms. Experimental results demonstrate an accuracy rate of over 85%, reducing manual verification efforts and improving consumer trust. The findings suggest that AI-driven fraud detection can significantly enhance the credibility of online reviews, fostering a more transparent digital marketplace. Keywords: Artificial intelligence, fake reviews, machine learning, sentiment analysis, e-commerce fraud