Inference of social media opinion trends in 2022 Italian elections

Simon Zollo, Matteo Cinelli (), Gabriele Etta (), Roy Cerqueti () and Walter Quattrociocchi ()
Additional contact information
Simon Zollo: UNIROMA - Università degli Studi di Roma "La Sapienza" = Sapienza University [Rome]
Matteo Cinelli: UNIROMA - Università degli Studi di Roma "La Sapienza" = Sapienza University [Rome]
Gabriele Etta: UNIROMA - Università degli Studi di Roma "La Sapienza" = Sapienza University [Rome]
Roy Cerqueti: GRANEM - Groupe de Recherche Angevin en Economie et Management - UA - Université d'Angers - Institut Agro Rennes Angers - Institut Agro - Institut national d'enseignement supérieur pour l'agriculture, l'alimentation et l'environnement, UNIROMA - Università degli Studi di Roma "La Sapienza" = Sapienza University [Rome]
Walter Quattrociocchi: UNIROMA - Università degli Studi di Roma "La Sapienza" = Sapienza University [Rome]

Post-Print from HAL

Abstract: Social media platforms play a significant role in political discourse, often serving as tools for political actors to disseminate partisan narratives, frequently encapsulated in concise slogans presented as hashtags. In this paper, we present a novel systematic framework leveraging network science tools and clustering algorithms to discern the political orientations of posts through their associated hashtags, that can be used in the context of opinion dynamics. Our results show that by applying this framework within the context of the 2022 Italian Elections, we successfully quantify the online activity of political coalitions and their supporters pre and post-election. By analyzing labeled posts derived from this framework we find a surge in user activity leading up to the election, followed by a pronounced decline afterward. Moreover, we note a remarkable shift in engagement toward the winning coalition post-election. Interestingly, at the coalition level, our findings reveal an inverse correlation between posting activity and the level of engagement received on social media platforms. Finally, a rank-size analysis of publication patterns among supporters during the pre-election period highlighted comparable trends in content generation across coalitions.

Keywords: Network science; Label propagation; Political elections; Opinion mining (search for similar items in EconPapers)
Date: 2025-04
Note: View the original document on HAL open archive server: https://univ-angers.hal.science/hal-05109451v1
References: Add references at CitEc
Citations:

Published in Expert Systems with Applications, 2025, 269, pp.126377. ⟨10.1016/j.eswa.2024.126377⟩

Downloads: (external link)
https://univ-angers.hal.science/hal-05109451v1/document (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:hal:journl:hal-05109451

DOI: 10.1016/j.eswa.2024.126377

Access Statistics for this paper

More papers in Post-Print from HAL
Bibliographic data for series maintained by CCSD ().