arrow
Return

Developing a Commenter Behavior-Based Framework for Characterizing YouTube Channels

delete2026-01-01
delete0
PRE
AI
S
Shadi Shajari
N
Nitin Agarwal *
DOI:10.1007/978-3-032-13513-1_9delete
deleteOriginal
deleteOriginal request for help
deleteShare
deleteSave
Abstract

Abstract

En 中文
As a major platform for global content dissemination, YouTube hosts thousands of channels covering diverse topics and attracting varied user engagement. Understanding the behavioral characteristics of these channels is essential for comparative analysis and broader platform insights. While channel-level activity can be studied through multiple signals, the comment section offers a particularly rich source of user interaction data. Patterns in commenter behavior provide an effective basis for characterizing channels and identifying similarities across them. This study introduces a framework for characterizing YouTube channels based on commenter activity, using a dataset of 70 channels that includes 711,301 videos, 13,221,243 commenters, and 129,653,609 comments and covers topics such as U.S. military affairs, geopolitical developments, news, and randomly selected content. The methodology combines co-commenter network construction, clique extraction, feature aggregation, pairwise feature analysis, clustering methods, majority voting, and Euclidean distance measurement to uncover patterns in channel behavior. We define five distinct characterizations based on key commenter activity features such as posting patterns, content diversity, and engagement structure. By relying on unsupervised methods and a combination of structural and behavioral features, this approach enables large-scale, label-free comparison of channels. It provides a flexible framework for identifying groups of channels with similar audience dynamics, offering insights that can support better channel characterization and future studies in social media behavior.
Keywords:
Social Network Analysis
Commenter Behavior
Majority Voting
Characterization
Euclidean distance
YouTube

Journal

S
SOCIAL NETWORKS ANALYSIS AND MINING, ASONAM 2025, PT I
IF:
0
Papers:
31
Citations:
0

Organization

U
University of Arkansas System
Scholars:
1.9W
Papers: 1.5W
Citations: 295
Cited Papers

Cited Papers

YouTube Data Collection Using Parallel Processing
err2020-05-01
err0
PREAI
errJoseph Kready; Shishila Awung Shimray; Muhammad Nihal Hussain; Nitin Agarwal
errShare
errSave
err
IF0
err
err0
PREAI
err
errShare
errSave
err
IF0
err
err0
PREAI
err
errShare
errSave
errShare
errSave
err
IF0
err
err0
PREAI
err
errShare
errSave
The expression of hate speech against Afro-descendant, Roma, and LGBTQ+ communities in YouTube comments
err2023-06-19
err0
errOAAI
errPaula Carvalho; Danielle Caled; Cláudia Silva; Fernando Batista; Ricardo Ribeiro
errShare
errSave
err
IF0
err
err0
PREAI
err
errShare
errSave
Thou Shalt Not Hate: Countering Online Hate Speech
err2019-07-06
err0
errOAAI
errBinny Mathew; Punyajoy Saha; Hardik Tharad; Subham Rajgaria; Prajwal Singhania; Suman Kalyan Maity; Pawan Goyal; Animesh Mukherjee
errShare
errSave
researcher View more