Inter- and Intrastate Network Analysis of COVID-19 Spread Using the Social Connectedness Index

Abstract

Since 2020, the outbreak of the coronavirus disease 2019 (COVID-19) pandemic has affected the entire world, and networks of human connections were identified as a factor that had potentially impacted the geographical spread of COVID-19. With the help of social media platforms, these networks have connected populations across the word and allowed people to view each other in close virtual proximity. Consequently, the Social Connectedness Index (SCI) is used to measure the strength of social connectivity across geographical regions through friendship ties on Facebook. The importance of social networks—and their relation to human connections—may correlate with the spread of COVID-19. Since these networks can have a potential effect on the spread of COVID-19, it is crucial to identify the factors that were associated with its spread during the pandemic. In order to analyze SCI data, a social network analysis was conducted to define the network parameters and perform calculations using graph theory. A correlation analysis was also performed to identify factors that correlated with the spread of COVID-19 cases using the data in the United States (US). Finally, the machine learning model was used to create a case prediction paradigm from the network parameters. The results showed that SCI can be used as a parameter to create a pandemic prediction model. Multiple linear regression also yielded satisfactory results that predicted the total number of positive cases measured by adjusted R2. In terms of the time frame, this study suggested that the parameters from the previous week can be used to predict the number of weekly infections. The findings showed that social networks had a greater impact on the prediction of current active cases than total positive cases. The social networks between counties within a state also held more importance than those across states.