Characteristics of two polarized groups in online social networks’ controversial discourse
Author(s)
Mahmoudi, Amin; Jemielniak, Dariusz; Ciechanowski, Leon
Download42001_2024_350_ReferencePDF.pdf (Embargoed until: 2025-12-09, 4.343Mb)
Publisher Policy
Publisher Policy
Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use.
Terms of use
Metadata
Show full item recordAbstract
In today’s interconnected world, online social networks play a pivotal role in facilitating global communication. These platforms often host discussions on contentious topics such as climate change, vaccines, and war, leading to the formation of two distinct groups: deniers and believers. Understanding the characteristics of these groups is crucial for predicting information flow and managing the diffusion of information. Moreover, such understanding can enhance machine learning algorithms designed to automatically detect these groups, thereby contributing to the development of strategies to curb the spread of disinformation, including fake news and rumors. In this study, we employ social network analysis measures to extract the characteristics of these groups, conducting experiments on three large-scale datasets of over 22 million tweets. Our findings indicate that, based on network science measures, the denier (anti) group exhibits greater coherence than the believer (pro) group.
Date issued
2024-12-30Journal
Journal of Computational Social Science
Publisher
Springer Nature Singapore
Citation
Mahmoudi, A., Jemielniak, D. & Ciechanowski, L. Characteristics of two polarized groups in online social networks’ controversial discourse. J Comput Soc Sc 8, 22 (2025).
Version: Author's final manuscript