Epidemic models for COVID-19 Information Diffusion in Mexican Social Media
Authors
Ismael Ariel Robles Martínez, Daniela Aguirre Guerrero, Carlos Joel Rivero Moreno, Gerardo Abel Laguna Sánchez
Abstract
Social media plays a pivotal role in the widespread sharing of information on the Coronavirus Disease 2019 (COVID-19). Social media platforms have been used across several countries by public administration as part of the communication plan in order to disseminate accurate information on COVID-19 and raising awareness in preventative behaviours.
However, social media platforms have also been used to spread misinformation and disinformation about COVID-19 (i.e. infodemic), which reduces the effective response of governments to the emergency of COVID-19 pandemic. This issue has raised concerns among many health organizations: the World Health Organization (WHO), has recognized that managing the COVID-19 infodemic is a critical part of controlling the COVID-19 pandemic.
The COVID-19 information diffusion in Mexico presents an interesting case study of polarized debate and infodemic on social media. On the one hand, the Mexican Government has deployed a COVID19 information campaign through daily conferences and social media. On the other hand, social media platforms exhibit infodemic and polarized opinions about the official health security measures.
This study aims to analyse the COVID-19 information diffusion in Mexico through a massive data analysis on Twitter and Reddit. First, Twitter and Reddit data related to COVID-19 in Mexico were retrieved through their respective API (Application Programming Interface) and using natural language processing. Then, an analysis of networks of hashtags and users over time and space was performed to characterize the information diffusion on the users networks. Finally, the information diffusion was fit in the epidemic model Susceptible-Infected-Recovered (SIR). As result, we have obtained different characterizations of the SIR model in Twitter and Reddit. We also found that polarized opinions about the same topic have also different patterns in the same social platform.
Keywords
Infodemic, CODIV-19, social media, social networks, epidemic models