Assessment, Implication, and Analysis of Online Consumer Reviews: A Literature Review

Abstract

The onset of e-marketplace, virtual communities and social networking has appreciated the influential capability of online consumer reviews (OCR) and therefore necessitate conglomeration of the body of knowledge. This article attempts to conceptually cluster academic literature in both management and technical domain. The study follows a framework which broadly clusters management research under two heads: OCR Assessment and OCR Implication (business implication). Parallel technical literature has been reviewed to reconcile methodologies adopted in the analysis of text content on the web, majorly reviews. Text mining through automated tools, algorithmic contribution (dominant majorly in technical stream literature) and manual assessment (derived from the stream of content analysis) has been studied in this review article. Literature survey of both the domains is analyzed to propose possible area for further research. Usage of text analysis methods along with statistical and data mining techniques to analyze review text and utilize the knowledge creation for solving managerial issues can possibly constitute further work.

Machine Learning in Detecting COVID-19 Misinformation on Twitter

“…For instance, Philander et al [2] used SA to understand public opinions towards hospitality firms. Moreover, the authors in [19] discussed the influence of SA on physical and online marketing, implication on product sales, and influence of pricing strategy. Shuyuan Deng et al [14] discussed the causality between sentiments and stock returns at an hourly level.…”

Text-rating review discrepancy (TRRD): an integrative review and implications for research

“…In addition, numerous surveys and consulting reports have suggested that, for a number of consumers and products (not applicable to all), consumer-generated reviews are valued more highly than reviews from 'experts' [21,61]. Therefore, Online Consumer Reviews (OCRs) can impact the consumer decision-making process to a greater extent than traditional media [1].…”

“…Reviewed papers are broadly first categorized into two themes: (1) research studies that built their models based on either rating scores or review texts, assuming that they were correlated implicitly with each other. In this part, we included research studies that have not considered any validation metrics to compute the relationship degree between available rating scores and other labelling techniques.…”

“…Since these scores are only computable using numerical ratings, text reviews should be converted into numerical values or star ratings. There are two ways to do this: (1) asking customers to express their opinions on products and services using star ratings or (2) calculate the overall ratings of the text reviews through the use of sentiment prediction techniques. Traditionally, it is assumed that user sentiments, which are expressed in the text reviews, should correlate highly with their score ratings [16,48].…”