Abstract
Using a novel methodological approach to measure emotions in Facebook comments, this Work in Progress (WIP) paper explores the relationship between negative feelings and ideological cross-posting behavior. Using the VoxPopuli data harvester, we collect over 770,000 public Facebook comments1 from the three major political campaign pages active during the Brexit referendum. After sorting users into ideological camps based on their reactions to campaign posts, we then examine their commenting patterns across ideological lines. Using three different methods of sentiment analysis, we identify negative and positive emotions and their fine-grained sub-categories in comments. The analysis reveals one quarter of all comments are cross-ideological posts, with Leave supporters overwhelmingly active in commenting on Remain posts. A comparison across the campaigns shows that Brexiteers are much more likely to express anger than Remainers.
Original language | English |
---|---|
Title of host publication | Proceedings of the 9th International Conference on Social Media and Society, SMSociety 2018 |
Publisher | Association for Computing Machinery (ACM) |
Pages | 246-250 |
Number of pages | 5 |
ISBN (Print) | 9781450363341 |
DOIs | |
Publication status | Published - 2018 Jul 18 |
Event | 9th International Conference on Social Media and Society, SMSociety 2018 - Copenhagen, Denmark Duration: 2018 Jul 18 → 2018 Jul 20 |
Conference
Conference | 9th International Conference on Social Media and Society, SMSociety 2018 |
---|---|
Country/Territory | Denmark |
City | Copenhagen |
Period | 2018/07/18 → 2018/07/20 |
Subject classification (UKÄ)
- Media Studies
Free keywords
- ACM proceedings
- Emotions
- Marketing
- NLP
- Political communication
- Sentiment analysis