From chunks to clusters: Identifying similarity features in social discussion

Choi, Yunseon

Permalink

https://hdl.handle.net/2142/113738 Copy

Description

Title

From chunks to clusters: Identifying similarity features in social discussion

Author(s)

Choi, Yunseon

Issue Date

2022-02-28

Keyword(s)

• Social media
• Online reviews
• Social discussion

Abstract

Users’ reviews on social media are crucial to understanding users’ interests and their opinions. Although there has been sufficient research on online reviews about products and services, there has been a lack of studies examining online reviews about books. This study extends our earlier work on frequency analysis of review words on online book reviews, which identified users’ interests in discussing books by analyzing the frequency of words users used in their book reviews on a social networking site. This paper intends to investigate whether the frequencies of the review words would represent similarities that would help understand the characteristics of books in selecting books for children. This study performs hierarchical cluster analysis on the selected books to identify homogeneous clusters of cases (books) based on selected characteristics (word frequencies). The finding of this study shows meaningful similarities in the social discussion by clustering books based on the characteristics of books. The results of this study help us understand the specific features of books and user behavior in discussing books on a social networking site. This study has implications for providing practical insights into the intrinsic values of users’ social discussion in identifying similarities among books.

Publisher

iSchools

Type of Resource

Other

Language

eng

Permalink

http://hdl.handle.net/2142/113738

Copyright and License Information

Copyright 2022 is held by Yunseon Choi. Copyright permissions, when appropriate, must be obtained directly from the author.

Owning Collections