Social Media Sentiment and COVID-19 Transmission: Results from a Time-varying SIR Model

COVID-19 has caused 2.48 million deaths worldwide. Efforts have been made to understand its transmission pattern and how to prevent its spreading. This is the first time the human race has experienced a pandemic with the prevalence of social media. Social media plays a critical role in disseminating COVID-19 related information. In this work-in-progress, we study the effect of social media on the transmission of COVID-19. Building upon the past literature on the role of social media in disaster management, we formulate a time-varying SIR model, namely SIRSM, to incorporate the interaction of the susceptible compartment and sentiment polarities in social media. Using Twitter data, we find that SIRSM can capture the spread of COVID-19 in the United States. In addition, our results indicate that negative sentiment on Twitter is associated with a lower transmission rate of the corresponding susceptible compartment