Examining differences among book reviews from various online platforms
| Published date | 11 November 2019 |
| DOI | https://doi.org/10.1108/OIR-01-2019-0037 |
| Date | 11 November 2019 |
| Pages | 1169-1187 |
| Author | Chengzhi Zhang,Tiantian Tong,Yi Bu |
| Subject Matter | Library & information science,Information behaviour & retrieval,Collection building & management,Bibliometrics,Databases,Information & knowledge management,Information & communications technology,Internet,Records management & preservation,Document management |
Examining differences among
book reviews from various
online platforms
Chengzhi Zhang and Tiantian Tong
Department of Information Management,
Nanjing University of Science and Technology, Nanjing, China, and
Yi Bu
Center for Complex Networks and Systems Research,
School of Informatics, Computing, and Engineering,
Indiana University, Bloomington, Indiana, USA
Abstract
Purpose –Websites have their own features in aspect preference (e.g. the relative importance platforms
place on product aspects in product evaluation). The purpose of this paper is to capture characteristics of
different book reviews on aspect preferences by opinion mining techniques.
Design/methodology/approach –The authors employ two indicators for identifying aspect preferences,
and propose a method for quantifying overall differences of reviews on aspect preferences through three
dimensions: aspect awareness, aspect satisfaction and comprehensive value.
Findings –The results show that bo ok reviews on e-commer ce websites contain inf ormation about
external aspects of a bo ok (e.g. hardcover), while those on socia l network websites pay more attention to
content-related asp ects of the book (e.g. stories). These results i ndicate that aspect preferences of reviews
vary from platforms and make it hard to evaluate book comprehensivel y based on single-sourc e
data. Online book reviews from a wide rangeof sources can assess book impact from multiple perspectives
and dimensions.
Practical implications –In order to illustrate the value of the authors’method, the authors show book
impact assessment bas ed on multi-source data as an applicatio n of these difference analyses. Furt hermore,
the authors present an ex ample of a book promotion to provid e customized marketing service s for different
user clusters.
Originality/value –This study investigates the influence of different data sources on book evaluation from
the content of book reviews. The authors also showcase potential applications of these analyses in book
impact assessment.
Keywords Sentiment analysis, Book evaluation, Aspect-level sentiment analysis,
Book impact assessments, Online book review, Review mining
Paper type Research paper
Introduction
Book reviews are a form of literary criticism in which books are analyzed based on their
content, style and originality (Editor, 2014). The quality of a book review varies from writer
to writer, but most of them provide an insightful view for the proto-reader (Gorraiz et al.,
2014). In the past, book reviews were mostly written by authors and editors, such as a “peer
review”process. With the development of Web 2.0, diversified book reviews generated by
users have appeared across the internet, such as on Goodreads (www.goodreads.cn) and
Amazon (www.amazon.cn). The amount of user-generated book reviews is huge, which is
several hundred times more than those based on peer reviews (Kousha et al., 2017).
Book reviews have played important roles in various disciplines. In linguistics, for
example, their structure and rhetorical content have been anatomized to describe the writing Online Information Review
Vol. 43 No. 7, 2019
pp. 1169-1187
© Emerald PublishingLimited
1468-4527
DOI 10.1108/OIR-01-2019-0037
Received 29 January 2019
Revised 21 May 2019
Accepted 5 August 2019
The current issue and full text archive of this journal is available on Emerald Insight at:
www.emeraldinsight.com/1468-4527.htm
This paper is funded by Major Projects of National Social Science Fund (No. 17ZDA291). The authors are
grateful to two anonymous reviewers for their insightful suggestions.
1169
Examining
differences
among book
reviews
style (Hartley, 2010). In library science, book reviews, as argued by Blake (1989), have been
used for developing and improving the quality of library collections. In business, the
relationships between sales and the volume of book reviews have been analyzed to
determine which metrics of online product review that data consumers depend on when
making their purchase decisions (Etzion and Awad, 2007). Book reviews have also been
studied in bibliometrics to fill the gap left by limited coverage of traditional metrics, such as
citation metric and library holdings (Zuccala and Guns, 2013). For example, Kousha and
Thelwall (2016) assessed the value of review ratings in choice for measuring the wider
impact of scholarly books. Subsequently, Zhou et al. (2016) conducted a sentiment analysis
method to mine rich aspect-level information in book online reviews so as to measure the
impacts of academic books.
Popular culture abounds that birds of a feather flock together, even in the online
commercial networks; different users with various motives select distinct platforms,
obviously, which makes platforms highlight different perspectives. For instance, three
popular platforms of book review, namely Amazon China, Dangdang, and Douban, have
differences according to their aspect frequency. As Table I shown, reviews on Amazon
China covering logistics and packaging made up more than 20 percent (15.7 and 5.5
percent, respectively); the number of logistics and packaging is up to 30 percent on
Dangdang (www.dangdang.com.cn); reviews on Douban (http://book.douban.com/)
contained more evidence about content (86.92 percent) and book type (6.11 percent).
Apparently, people will assign different weights to the aspects based on their information
acquisition and processing activities in product evaluation, which is said to be the aspect
preference (Liu and Karahanna, 2017). As platforms have their own features in aspect
preferences, single data source may affect the generalizability of the evaluation results.
Understanding the difference among platforms can contribute to reveal the influence of
different data sources on book evaluation so as to better incorporate various book online
reviews in assessment exercises. However, none of the existing literature, as far as we
know, has considered the differences among platforms when studying the possibility of
assessing books by online reviews.
There have been few studies that examine whether reviews vary across platforms and if
so, in what ways the reviews become different. Therefore, the current paper aims to examine
the disparity among platforms by analyzing linguistics of book online reviews. This study
selects three book review platforms: Amazon China, Dangdang, and Douban, and considers
the popularity on the internet and types of platforms (e.g. social network websites and
e-commerce websites). Ultimately, we can portray a comprehensive picture of characteristics
of various platforms and expect to show their disparity on aspect preferences.
To achieve our goal, we employ two measurements that identify aspect preferences,
namely aspect awareness and aspect satisfaction, which have been widely adopted to
quantify consumers’psychology in marketing (Bailey, 2005; Cronin et al., 2000). In this
study, aspect awareness refers to the amount of information about aspect-level performance,
which is often unevenly distributed across aspects. Aspect satisfaction is a measure of how
products and services meet/surpass customer expectation; it reflects the conflict degree of
the same aspect. Moreover, for highlighting the overall differences of reviews on aspect
Percentage of aspect frequency
Platform Content (%) Type (%) Version (%) Logistics (%) Packaging (%)
Amazon China 70.6 3.6 4.7 15.7 5.5
Dangdang 64.61 1.75 2.77 28.73 2.15
Douban 86.92 6.11 6.93 0.12 0.05
Table I.
Proportion of aspects
in three platforms
1170
OIR
43,7
To continue reading
Request your trial