Social Media Analysis¶
Empirical study of user behavior, content patterns, and information dynamics on social media platforms. This includes comparative platform analysis, engagement measurement, content moderation effects, and quantitative characterization of how information spreads across different network architectures.
Key datasets¶
- FUBU-EPSTEIN — 54.38M Twitter statuses (2019–2023) with 37M-edge social contact graph; continuous near-real-time collection capturing deleted/edited posts and user profile evolution; enables longitudinal diffusion and network analysis
Key papers¶
- Suarez-Lledo & Alvarez-Galvez (2021) — Systematic review of 69 studies characterizing health misinformation prevalence across platforms; demonstrates how different social media platforms attract different analytical approaches (Twitter → network analysis, YouTube → content evaluation), revealing platform-specific affordances for misinformation research
- Zannettou et al. (2018) — On the Origins of Memes by Means of Fringe Web Communities: Large-scale comparative analysis of 160M images across Twitter, Reddit, /pol/, and Gab using perceptual hashing and clustering; quantifies influence between communities using Hawkes processes
- Memon & Carley (2020) — Characterizing COVID-19 Misinformation Communities — Network and sociolinguistic analysis of competing Twitter communities spreading COVID-19 misinformation vs. fact-checking; measures network density, bot prevalence, and linguistic patterns (LIWC) to characterize community structure
- Cinelli et al. (2020) — The COVID-19 Social Media Infodemic — Multi-platform comparative analysis of misinformation diffusion across Twitter, Instagram, YouTube, Reddit, and Gab during COVID-19 pandemic using epidemic models
- Cinelli et al. (2021) — The echo chamber effect on social media — Analyzes information diffusion patterns and algorithmic sorting across platforms
- Perceived and Intended Sarcasm Detection with Graph Attention Networks — Sarcasm detection via graph attention modeling user history and social network structure on Twitter; distinguishes between perceived and intended sarcasm