Fake Review Detection Using Machine Learning and Deep Learning, Distilled-Bert

Fake Review Detection Using Machine Learning and Deep Learning, Distilled-Bert

Mayur Kadam, Shubham Marewad, Chetan Nemade, Parikshit Mote

Department Computer engineering, ZCOER, Pune, Maharashtra, India

ABSTRACT

In today's digital economy, the reliability of internet reviews is crucial in influencing consumer behavior. However, spam reviews—deceptive or manipulative entries that blur real customer feedback—have become a problem due to the extensive use of user-generated content across social media and e-commerce platforms. Although the nature and implications of such reviews have been examined in previous studies, more sophisticated detection methods are required due to the growing sophistication of spamming techniques. Using natural language processing (NLP) models that can recognize contextual inconsistencies and hidden patterns in review content, this paper explores the use of contemporary machine learning and deep learning approaches for spam review detection[9]. To improve model accuracy, behavioral characteristics like reviewer credibility, posting patterns, and temporal activity are also looked at.The goal of this research is to create a strong, scalable solution to stop spam reviews and preserve the integrity of online feedback systems by combining textual and behavioral insights. The study also underscores how social media is increasingly influencing consumer choices and how sophisticated sentiment analysis tools can offer more profound understandings of consumer intent, ultimately promoting equitable and reliable online marketplaces[5].

Keywords: Fake Reviews, Nueral Network, machine learning, deep learning, Behaviour Analysis, Text Analysis,  Natural Language Processing,  Genuine Reviews,  E-Commerce Multi Model System