Finding “just right” books for children: analyzing sentiments in online book reviews
| DOI | https://doi.org/10.1108/EL-01-2019-0018 |
| Pages | 563-576 |
| Published date | 03 June 2019 |
| Date | 03 June 2019 |
| Author | Yunseon Choi |
| Subject Matter | Information & knowledge management,Information & communications technology,Internet |
Finding “just right”books for
children: analyzing sentiments in
online book reviews
Yunseon Choi
Department of Library and Information Studies, Valdosta State University,
Valdosta, Georgia, USA
Abstract
Purpose –This study is part of a larger researchproject which aims to investigate whether sentiments in
online reviews onchildren’s books would represent significant factors which are useful for selectingthe right
books for children. This paper aims to examine whether positive, negative or neutral attitude would be
directlyassociated with the overall ratings of books.
Design/methodology/approach –The study investigatessubjectivity and polarity of online reviews on
children’s books such as neutral, positive or negative sentiment. For the investigation of a statistical
association between the sentimentvalues and the rating scores, this study performs correlation analysis. For
a clear explanation of the factorsaffecting the relationships between the sentiment value and the rating score,
this study uses the concept-levelsentiment analysis of online reviews.
Findings –The findings of this study demonstrate that there is a weak or low correlation between the
sentiment valueand the rating score of a book and they arehardly related for most books. Theresults of this
study alsouncover key contributing factorsthat affected the correlationsbetween two variablesand made the
relationshipweak.
Research limitations/implications –This study increases awareness of the implications of online
reviewsas user-generated contents for complementingthe existing controlled vocabulary.
Practical implications –This study contributes to improving library catalogs by using latent topics
extractedfrom online reviews which provide additional access points for assistingin the selection of books.
Originality/value –Althoughseveral studies have conducted on onlinereviews in the domain of business,
no research appears to exist on the sentiment analysisof online reviews about children’s books. This study
attempts to address the potential and challengesassociated with using online reviews to help find the right
books for children.
Keywords Online reviews, Sentiment analysis, Social media, User-generated content
Paper type Research paper
Introduction
Sentiment analysis, known as opinion mining and subjectivity analysis, is analyzing people’s
opinions and attitudes from written reviews to rate products or services (Liu, 2010). With the
increasing popularity of online reviews at social networking sites, online reviews have become
a primary information source for a decision before making a purchase (Chen et al., 2018;Liu,
2006;Park et al., 2007;Sparks, 2013). User reviews or opinions are helpful not only for other
users to make decisions about purchasing a product but also for the company to know about
the positive and negative features of their product (Raju et al., 2016). Several studies on
sentiment analysis have been conducted on online reviews in the domain of business, such as
The research was partially supported by a Faculty Research Seed Grant from the Office of Sponsored
Programs and Research Administration at Valdosta State University.
Analyzing
sentiments in
online book
reviews
563
Received30 January 2019
Revised10 April 2019
Accepted13 May 2019
TheElectronic Library
Vol.37 No. 3, 2019
pp. 563-576
© Emerald Publishing Limited
0264-0473
DOI 10.1108/EL-01-2019-0018
The current issue and full text archive of this journal is available on Emerald Insight at:
www.emeraldinsight.com/0264-0473.htm
To continue reading
Request your trial