What are the diffusion patterns of COVID-19 virus spread, based on SIRsim and SIRemp?

• SIMsim shows only 5% of Infected and 0.19% of Removed cases compared to world population
• Infection and deaths increase logarithmically
• Differences in simulated and actual models found – SIMemp distribution is relatively flat compared to increasing SIRsim distribution
• Due to not considering preventative measures in models

What are the diffusion patterns of information cascades on Twitter (INFOcas), with respect to retweets, quote tweets, and replies?

• All 3 cascades displayed linear-log distributions, specifically power law decays
• Retweets have fastest decays of 4 hours; quote tweets 3 days; reply 2 days
• While retweets are most numerous, the length of interaction with original tweets is much shorter than the interactions between quote tweets and replies

What are the major differences in diffusion patterns between SIRsim, SIRemp, and INFOcas?

• Infected cascades: strong positive correlation between SIRsim and INFOcas retweets and quote tweets; weak correlation between SIRemp and all 3 INFOcas cascades
• Removed cascades: strongest correlations between each of the INFOcas cascades
• COVID-19 tweets get retweeted quickly, then quoted and replied