From Fake-Review Detection to Verifiable Reputation Infrastructures: A Computational Mapping of E-Commerce Review Governance Using BERTopic


Sağlam M. H., Aygören O.

Informatics, cilt.13, sa.11, ss.1-34, 2026 (Scopus)

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 13 Sayı: 11
  • Basım Tarihi: 2026
  • Doi Numarası: 10.3390/informatics13110202
  • Dergi Adı: Informatics
  • Derginin Tarandığı İndeksler: Scopus
  • Sayfa Sayıları: ss.1-34
  • Yıldız Teknik Üniversitesi Adresli: Evet

Özet

Manipulated reviews undermine the credibility of e-commerce reputation systems, yet research on detection, verification, and governance remains fragmented. This review maps the thematic and temporal structure of a defined journal corpus on e-commerce review integrity and develops a framework for evaluating verifiable reputation infrastructures. A February 2026 search of the Web of Science Core Collection and independent screening, reported using PRISMA 2020, yielded 212 English-language journal articles and reviews published between 2010 and 2025. Titles, abstracts, and keywords were analyzed using BERTopic. Five themes emerged: platform governance and enforcement; fake-review detection and adversarial methods; verified purchase and identity assurance; blockchain and smart-contract reputation design; and trust, privacy, and regulatory compliance. Within the retrieved corpus, later publication periods included research on participation verification and programmable reputation design alongside detection. The synthesis integrates the five themes into a framework that specifies credibility, deterrence, fairness, verifiability, and implementation feasibility as criteria for evaluating e-commerce reputation infrastructures. The framework positions blockchain as a supporting infrastructure whose potential value depends on reliable transaction evidence, enforceable rules, and accountable oversight. The review advances a governance-oriented understanding of review integrity and specifies priorities for evaluating how detection, verification, and institutional safeguards operate together.

Manipulated reviews undermine the credibility of e-commerce reputation systems, yet research on detection, verification, and governance remains fragmented. This review maps the thematic and temporal structure of a defined journal corpus on e-commerce review integrity and develops a framework for evaluating verifiable reputation infrastructures. A February 2026 search of the Web of Science Core Collection and independent screening, reported using PRISMA 2020, yielded 212 English-language journal articles and reviews published between 2010 and 2025. Titles, abstracts, and keywords were analyzed using BERTopic. Five themes emerged: platform governance and enforcement; fake-review detection and adversarial methods; verified purchase and identity assurance; blockchain and smart-contract reputation design; and trust, privacy, and regulatory compliance. Within the retrieved corpus, later publication periods included research on participation verification and programmable reputation design alongside detection. The synthesis integrates the five themes into a framework that specifies credibility, deterrence, fairness, verifiability, and implementation feasibility as criteria for evaluating e-commerce reputation infrastructures. The framework positions blockchain as a supporting infrastructure whose potential value depends on reliable transaction evidence, enforceable rules, and accountable oversight. The review advances a governance-oriented understanding of review integrity and specifies priorities for evaluating how detection, verification, and institutional safeguards operate together.