Abstract
Consumers often rely on online reviews and opinions posted on social media to make a decision when they purchase products or services. This article addresses what are collectively referred to as opinion spam, which are opinions posted by fake reviewers who seek to promote or tear down target entities for financial gain. This has led industry and academic research to seek to develop an efficient and scalable framework to detect such opinion spam. Yelp dataset for online reviews are studied using graph-based methods that leverage the relational ties among reviewers, reviews, and businesses. Yelp user networks is considered, in which reviewer nodes are connected to each other as friends relationship. We investigate structural properties of user networks for recommended (non-spam) and fake (spam) reviewer groups. It has demonstrated that networks for groups of recommended reviewers show characteristics of a small-world network. However, networks for groups of fake reviewers reveal properties closer to those of a random network. Clues from the study of structural properties of user networks are used to extend a fraud detection framework which exploits network effects among reviewers and products. Our extended framework is more effective on detecting frauds in Yelp review dataset than previous works.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.