Identifying the topology of the K‐pop video community on YouTube: A combined Co‐comment analysis approach

Abstract

YouTube is a successful social network that people use to upload, watch, and comment on videos. We believe comments left on these videos can provide insight into user interests, but to this point have not been used to map out a specific video community. Our study investigates whether and how user commenting behavior impacts the topology of the K-pop video community through analysis of co-commenting behavior on these videos. We apply a traditional author cocitation analysis to this behavior, in a process we refer to as co-comment analysis, to detect the topology of this community. This involves: a) an analysis of user co-comments to elicit the inclination of user homophily within the community; b) an analysis of user co-comments, weighted frequency of co-comments, to detect user interests in the community; and c) an analysis of user co-comments, weighted sentiment scores, to capture user opinions by polarity. The results indicate that users who comment on specific K-pop videos also tend to comment on topically similar YouTube videos. We also find that the number of comments made by users correlates with the degree of positivity of their comments. Conversely, users who comment negatively on K-pop videos are not inclined to form specific user groups, but rather present only their opinions individually.

The ‘Guggentube’ Phenomenon: Breaking the Boundaries of a ‘Digital Museum’ Space

“…M any scholars have used digital ethnography to examine user opinions in various online communities, ranging from highly political communities to social groups or fan communities (Khalikova and Fish 2016; Tuwei and Tully 2017). They document the effective ness of this method to 'reveal explicit user perspectives and community interests' (Song et al 2015(Song et al , p.2581 M y use of digital ethnography is based on an analysis of both the content and language employed in YouTube users' textual practices. This method was instrumental to explore how YouTube Play audiences understood a contemporary museum and arts, how they experienced 'digital museum' spaces and what they thought or felt the YouTube Play project in particular (Grincheva 2017;.…”

Discovery of Multi-Objective Overlapping Communities within Social Networks using a Socially Inspired Metaheuristic Algorithm

“…Song et al have introduced a novel approach for identifying the topological community structure for K-pop videos on YouTube through analysis of co-commenting behavior on these videos utilizing the adapted co-link analysis and author co-citation [17].…”

Cutting Through the Comment Chaos: A Supervised Machine Learning Approach to Identifying Relevant YouTube Comments

“…It has been shown, for example, that people’s attitudes towards and experiences of videos depend on the comments that accompany these videos (e.g., Hsueh et al, 2015; Möller et al, 2021; Shi et al, 2014; Waddell & Sundar, 2017; Walther et al, 2010; Ziegele et al, 2018). Other studies focused on how viewers use comments to converse about the content that they are watching (e.g., Dubovi & Tabak, 2020; Poché et al, 2017; Song et al, 2015).…”

“…Understanding the extent to which YouTube comments reflect viewers’ reactions to or experiences of videos is a necessary step if we want to advance our theoretical knowledge about online interpersonal communication on social media platforms. For example, various content analyses of YouTube comments detected the sentiment of comments to learn more about viewers’ opinions and attitudes (e.g., Siersdorfer et al, 2010; Song et al, 2015; Thelwall, Sud, et al, 2012). However, if we assume that large amounts of comments are irrelevant and not related to the respective video, sentiment analysis might falsely categorize these comments as meaningful and might draw a biased picture of the public opinion on the video.…”