Skip to main navigation Skip to search Skip to main content

ALL THAT GLITTERS IS NOT GOLD: IDENTIFICATION, EXTRACTION, AND APPLICATIONS OF PRODUCT INFORMATION FROM ONLINE CONSUMER REVIEWS

Research output: Contribution to journalReview articlepeer-review

Abstract

Existing research showed that voting the helpfulness of online reviews may not be the most appropriate measure of the quality of the information provided by those reviews. This paper suggests a new approach to rank online consumer reviews based on three variables of the relative information content of the textual portion of each review and the readability of that text, as reflecting how well the information is conveyed. The study shows the value of this approach over and above previous measures based on human votes. The study then extends that approach to the evaluation of the contribution of individual reviewers. The methodology was tested on 49,143 online reviews from three dissimilar product categories available on Amazon.com: appliances, computers, and luxury beauty. The implications of the proposed methodology and its potential applicability are discussed.

Original languageEnglish
Pages (from-to)312-334
Number of pages23
JournalJournal of Electronic Commerce Research
Volume27
Issue number4
StatePublished - Nov 2026

Keywords

  • e-Commerce
  • e-WOM
  • Information entropy
  • Natural language processing
  • Online consumer reviews
  • Review helpfulness
  • Reviewer expertise

Fingerprint

Dive into the research topics of 'ALL THAT GLITTERS IS NOT GOLD: IDENTIFICATION, EXTRACTION, AND APPLICATIONS OF PRODUCT INFORMATION FROM ONLINE CONSUMER REVIEWS'. Together they form a unique fingerprint.

Cite this