Published in Natural Hazards Review, along with co-authors Liangdong Lu, Jia Xu, and Jiuchang Wei: “Quantifying the Psychological Online Communities Considering the Relationship between COVID-19-Related Threat, Information Uncertainty, and Risk Perception.”

Abstract: This study employed deep learning to analyze a substantial data set of 109.13 million COVID-19-related microblogs, leading to the construction of a specialized risk perception indicator dictionary. Employing this dictionary, we were able to capture the dynamic fluctuations in risk perception within online communities across various cities in real time. This approach highlighted the varying intensities of public response to the evolving crisis during the isolation and normalization stages of the pandemic. We observed that COVID-19-related transmission threat and information uncertainty significantly influenced public risk perception at different stages of the pandemic. Innovatively, our study quantifies public psychological resilience within online communities by examining the equilibrium between public risk perception and objective COVID-19-related risks. This equilibrium is conceptualized as the alignment of public perception with the evolving reality of COVID-19 threat and information. We investigated psychological resilience in two dimensions: adaptability, indicated by the extent of deviation from this equilibrium, and agility, reflected in the rate at which equilibrium is reestablished. Our study not only unveils new insights into the intricate relationship among public risk perception, the evolving risks, and psychological resilience but also offers empirical evidence to inform risk management strategies in online communities at different stages of a crisis.