Tweet analysis refers to computational methods for extracting, classifying, and interpreting information from Twitter posts. Methods in this area address stance classification, sentiment and emotion recognition, bot detection, rumor identification, and credibility assessment—all critical for understanding information propagation and misinformation spread on social media.
Tasks and applications
- Stance detection: Classifying whether a tweeter supports, opposes, or is neutral toward a claim or entity
- Sentiment analysis: Measuring positive, negative, or neutral sentiment expressed toward a topic
- Rumor tracking: Identifying and following the spread of unverified claims
- Misinformation detection: Detecting false or misleading claims in tweets
- Bot detection: Identifying automated accounts and coordinated inauthentic behavior
- Influence and amplification: Measuring how tweets spread through retweets and engagement
- Argument extraction: Mining arguments and evidence from tweet threads
Key papers
- 2016-mohammad-stance-detection-tweets — SemEval-2016 Task 6 establishing the first large-scale shared task on stance detection in tweets; 4,163 annotated tweet-target pairs; foundational benchmark for the field
Related topics
- sentiment-analysis — broader task encompassing emotions and opinions in text
- misinformation-detection — identifying false claims, which often spread via tweets
- social-media-analysis — studying communication patterns and information flow