Detecting fake-review buyers using network structure: Direct evidence from Amazon

Sherry He, Brett Hollenbeck, Gijs Overgoor, Davide Proserpio and Ali Tosyali
Additional contact information
Sherry He: a Anderson School of Management, University of California, Los Angeles, CA 90095;
Gijs Overgoor: b Saunders College of Business, Rochester Institute of Technology, Rochester, NY 14623;
Davide Proserpio: c Marshall School of Business, University of Southern California, Los Angeles, CA 90089
Ali Tosyali: b Saunders College of Business, Rochester Institute of Technology, Rochester, NY 14623;

Proceedings of the National Academy of Sciences, 2022, vol. 119, issue 47, e2211932119

Abstract: Online reviews significantly impact consumers’ decisions and are seen as crucial to the success of online markets. Despite this, the prevalence of fake reviews is arguably higher than ever, despite two decades of academic research on identifying and regulating them. We use data in which we directly observe which products buy fake reviews, and study how to identify them. We show that products buying fake reviews are highly clustered in the product reviewer network, due to their reliance on common reviewers. This allows us to detect them with high accuracy using both supervised and unsupervised methods. Unlike approaches relying on reviews’ text, this approach is more robust to manipulation by sellers. Moreover, it is scalable and generalizable to many settings.

Keywords: online reviews; networks; machine learning; text analysis (search for similar items in EconPapers)
Date: 2022
References: Add references at CitEc
Citations:

Downloads: (external link)
http://www.pnas.org/content/119/47/e2211932119.full (application/pdf)

Related works:
Working Paper: Detecting fake review buyers using network structure: Direct evidence from Amazon (2024)
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:nas:journl:v:119:y:2022:p:e2211932119

Access Statistics for this article

More articles in Proceedings of the National Academy of Sciences from Proceedings of the National Academy of Sciences
Bibliographic data for series maintained by PNAS Product Team ().