As COVID-19 continues, social media platforms such as Facebook have become an increasingly important tool for communication and information sharing for public and government agencies. The generic disaster management cycle (mitigation, preparedness, response, and recovery) provides systematic guidance to the public and government agencies to respond to the crisis and suggest appropriate measures for different disaster stages. In this study, we examine various trending topics and themes during the COVID-19 outbreak. Using this generic disaster management cycle as our guiding framework, we examine news topics' evolution during the COVID-19 pandemic on Facebook during each of the four phases. Guided Latent Dirichlet Allocation (Guided LDA) is used for topic modeling to identify topics and themes, and text network analytics is used to understand the connectedness of these news topics during each phase and their evolution.

Abstract Only