This paper presents a comprehensive network analysis of medical influencers on social media, examining how their identities and discourse patterns affect health information dissemination and public trust during health crises.

Abstract: Health misinformation spread unchecked during COVID-19, complicating public-trust efforts by authorities. This work compiles tweets from the top 100 medical influencers by “Influencer Score” and builds a few-shot, multi-label classifier for analyzing their discourse patterns.

Key Contributions:

• Network representation of medical influencer identities
• Multi-label classification with 70% F1 score
• Thematic analysis of health discourse patterns
• Dynamic knowledge graph construction for discourse networks

Recommended citation: Z. Guo, E. Simpson, and R. Bernardi, “Medfluencer: A network representation of medical influencers’ identities and discourse on social media,” in epiDAMIK 2024: The 7th International Workshop on Epidemiology meets Data Mining and Knowledge Discovery at KDD 2024.