Claimspotting. On the Role and Automation of Fact-checking
Veröffentlichungsdatum
2025-02-24
Autoren
Nenno, Sami
Betreuer
Humprecht, Edda
Gutachter
Zusammenfassung
The present PhD framework text summarises six contributions that were written between 2022 and 2024, some of which have been published in peer-reviewed journals. The subject of the PhD thesis is claim detection, which is the automated monitoring of potential misinformation to assist fact-checkers. The two main research questions are how to design claim detection with respect to the needs of fact-checkers and the behaviour of social media users, and how to implement it with regard to technical limitations. Accordingly, the contributions are either empirical studies or methodological contributions. In particular, they include a quantitative analysis of fact-checks, interviews with fact-checkers, a computer-assisted analysis of flagged content on Facebook, the evaluation of existing claim detection datasets, the introduction of a new dataset for it, and the development of a domain specific named entity recognition model. The framework text provides an introduction to fact-checking in Germany, which is followed by an overview of the contributions. The main part introduces and answers the research questions; the remainder places claim detection in the overarching context of fact-checking and misinformation.
Schlagwörter
Fact-checking
;
Disinformation
;
Natural Language Processing
Institution
Fachbereich
Dokumenttyp
Dissertation
Sprache
Englisch
Dateien![Vorschaubild]()
![Vorschaubild]()
Lade...
Name
Nenno_Claimspotting_Dissertation.pdf
Size
4.68 MB
Format
Adobe PDF
Checksum
(MD5):dc7aab4246b1a2b7209fc9ff34cd5110
Lade...
Name
Nenno_Claimspotting_Dissertation.pdf
Size
4.68 MB
Format
Adobe PDF
Checksum
(MD5):dc7aab4246b1a2b7209fc9ff34cd5110
