This paper provides the first comprehensive definition of social media bots and analyzes their behavioral differences from humans.
This study is based on a large-scale analysis of social media tweets across ∼ 200 million users across 7 events
-----
https://arxiv.org/abs/2501.00855
Methods in this Paper 🔍:
→ Develops a first-principles definition: "A Social Media Bot is an automated account that carries out mechanics for content creation, distribution, and relationship formation"
→ Analyzes bot behavior across 7 major events using 5 billion tweets from 200 million users
→ Compares bots vs humans on linguistic patterns, identity presentation, and network structure
-----
Key Insights 💡:
→ Bots make up 20% of social media users, spiking to 43% during political events
→ Bots use more automated features (hashtags, mentions) while humans use more cognitive features (replies, media)
→ Bots have star-shaped networks optimized for information spread, humans have hierarchical networks
-----
Results 📊:
→ Bots have 8.33% denser communication networks than humans
→ Only 21.4% of bots present clear identities vs 27% of humans
→ Bots interact primarily with humans (90.34% of interactions)
------
Are you into AI and LLMs❓ Join my daily AI newsletter. I will send you 7 emails a week analyzing the highest signal AI developments. ↓↓
🎉 https://rohanpaul.substack.com/
Share this post