Relationship between filter bubbles and the sharing of fake news on social media: an empirical study on Twitter/X
Files
Garcia_09432000_Labar_18012000_2025.pdf
Open access - Adobe PDF
- 4.52 MB
Details
- Supervisors
- Faculty
- Degree label
- Abstract
- Over the last few years, the rise of social media has revolutionised the way information is consumed. This transformation offers many advantages, but it also has its drawbacks. This thesis focuses on two major concerns: the sharing of fake news and the enclosure of users in filter bubbles on social media. Drawing on a set of tweets posted by 864 users who shared either fake or real news according to the Gossip Cop website, we investigate four research questions: Are users who share fake news exposed to less diverse content? Do they tend to have stronger opinions? Do they tend to share more recommended content rather than original tweets? Are there any features that distinguish this type of user? Using a quantitative approach, we combine topic modelling and Rao’s diversity measure to assess content diversity, sentiment analysis to evaluate intensity of opinions, and statistical comparisons to analyse user publication behaviour and explicit features. Our results show that users who share fake news are significantly more likely to be trapped in a filter bubble – as they are faced with less thematically diverse content – and to repost recommended content through retweets and quoted tweets, suggesting stronger algorithmic influence. Additionally, these users tend to be more active on the platform in terms of posting, liking, and following, and their posts tend to generate more retweets. However, no evidence suggests that their opinions are more extreme. These findings contribute to a better understanding of the relationship between the spread of fake news and the filter bubble phenomenon on social networks. They highlight the need to place greater emphasis on diversity in algorithms and performance indicators, and to invest more in preventive detection strategies and user awareness to mitigate the risks posed by these two phenomena.