Journal of King Saud University - Computer and Information Sciences | 2021
Research on false review detection Methods: A state-of-the-art review
Abstract
Abstract Fake reviews are popular today where product reviewers write the reviews without experiencing or purchasing the product on e-commerce and restaurant portals. Currently, the false review recognition method uses the systematic review process to extract, summarise, and classify the meaningful content of the research, compare and analyse the representation power of various false attributes, and the recognition method s performance. Feature design and recognition method design are the key steps for false review text recognition. The procurement of a large-scale labelled review dataset is difficult in recent research. They were only identifying fake review texts used as the core of the discussion. The article presents an assessment of fake reviews detection in different domains (hotels and e-commerce). In this article, we have also identified the relation between fake reviewers and groups of fake reviewers. We have analysed and pointed out the existing research problems in data acquisition, false feature design, and recognition method design to suggest future research on false review detection.