Authors : Meera Sawalkar; T. M. Mane; Rushikesh Kenjale; Tushar Gadekar; Jayprakash Jadhav; Atharva More; Atharva Phadatare

Volume/Issue : Volume 10 - 2025, Issue 6 - June

Google Scholar : https://tinyurl.com/45zk7nf5

DOI : https://doi.org/10.38124/ijisrt/25jun429

Note : A published paper may take 4-5 working days from the publication date to appear in PlumX Metrics, Semantic Scholar, and ResearchGate.

Abstract : In the digital age, online reviews significantly influence consumer behavior and business reputation. However, the rise of fake and irrelevant reviews has become a growing concern, affecting the credibility of e-commerce platforms and misleading potential customers. This research presents a comprehensive AI/ML-based system designed to detect and filter out such deceptive content. The system leverages natural language processing (NLP) techniques and a Markov Chain Model to identify gibberish or contextually irrelevant text. It also integrates image verification by analyzing EXIF metadata to validate whether submitted review images are authentic and correspond to the product in question. By combining structured review analysis with machine learning models and metadata validation, the system enhances the reliability of online feedback. This solution not only helps e-commerce platforms maintain content integrity but also builds greater trust among users, ultimately supporting more informed purchasing decisions.

Keywords : NLP, AI-ML, Review Monitoring, Fake Review Detection, Spam Review Detection, Markov Model.

References :

1. Abhijeet A Rathore, “Fake Reviews Detection Using NLP Model and Neural Network Model” (IJERT 2023)
2. Cheolgi Kim, “Efficient Detection of Irrelevant User Reviews Using Machine Learning” (MDPI 2024)
3. Rami Mohawesh, “Fake Reviews Detection: A Survey”, (IEEE 2021)
4. Wesam Hameed Asaad, “Fake Review Detection Using Machine Learning” (IIETA 2023)
5. Miroslav Ölvecký, “Digital image forensics using EXIF data of digital evidence” (ICETA 2021)