Debunking Characterization of Fact-checking Articles
Name
Peeter Niidas
Abstract
Nowadays, as we are facing fast growth and propagation of different information, it is vital to understand the characteristics of fact-checking articles to combat disinformation. This thesis aims to investigate debunking the characterisation of fact-checking articles by analysing their semantic and linguistic properties and social media influence. The thesis investigates the influence of negations, emotions, multilingualism and social media by using text analysis tools such as regex and multiple NLP libraries. Social media analysis goes deeper by detecting and analysing the dataset’s URLs.
The results show some role of negations, emotions and overall tonality in fact- checking articles. The results shed light on the influence of mentioned linguistic and semantic features on their effectiveness. The thesis also investigates the influence of emotions and social media mentions of fact-checking articles. The thesis aims to provide preliminary information hidden in these connections and guide future research to improve understanding of content in fact-checking articles.
For future guidelines, this thesis proposes developing automated tools, improving media literacy education, exploring cross-platform analysis and evaluating the impact of fact-checking strategies. These strategies could enhance understanding of the situation of misinformation and contribute to developing a more effective fact-checking methodology. As most of the research in this field is fractured and every research team has followed their methodology, this thesis emphasises the importance of developing standardised techniques to speed up future research and achieve comparability.
The results show some role of negations, emotions and overall tonality in fact- checking articles. The results shed light on the influence of mentioned linguistic and semantic features on their effectiveness. The thesis also investigates the influence of emotions and social media mentions of fact-checking articles. The thesis aims to provide preliminary information hidden in these connections and guide future research to improve understanding of content in fact-checking articles.
For future guidelines, this thesis proposes developing automated tools, improving media literacy education, exploring cross-platform analysis and evaluating the impact of fact-checking strategies. These strategies could enhance understanding of the situation of misinformation and contribute to developing a more effective fact-checking methodology. As most of the research in this field is fractured and every research team has followed their methodology, this thesis emphasises the importance of developing standardised techniques to speed up future research and achieve comparability.
Graduation Thesis language
English
Graduation Thesis type
Master - Conversion Master in IT
Supervisor(s)
Shakshi Sharma, Rajesh Sharma
Defence year
2023