Locked down in Wuhan: Exploring the evolution of public emotions during the COVID-19 pandemic (Preprint)

Abstract

BACKGROUND In late December 2019, Wuhan Municipal Health Commission reported the first cases of SARS-CoV-2, the underlying virus that caused the devastating outbreak of the coronavirus (COVID-19). On 23 January 2020, the government of Wuhan announced a city-wide lockdown with the aim of controlling the spread of the virus. With outbreaks of COVID-19 around the world, lockdown restrictions are routinely imposed to limit the spread of the virus and reduce the strain on healthcare services. During periods of lockdown, social media has become the main channel for citizens to exchange information and communicate with friends and family. Public emotions and opinions are being generated and shared rapidly online with citizens using internet platforms to reduce anxiety and stress, and stay connected while isolated. OBJECTIVE This study aims to explore the regularity of emotional evolution by examining public emotions expressed in online discussions during the lockdown of Wuhan in January 2020. We divide the lockdown into different phases and analyze the distribution of emotions against different dimensions. Further, the temporal evolution of emotion and the topic-based emotional distribution during each phase of the Wuhan lockdown is determined. METHODS Data related to the Wuhan lockdown was collected from Sina Weibo, the most active microblogging site in China, by web crawler. In this study, the Ortony, Clore, and Collins (OCC) model, Word2Vec, and Bi-directional Long Short-Term Memory model were employed to determine emotional types, train vectorization of words, and identify each text emotion for the training set. Latent Dirichlet Allocation models were also employed to mine the various topic categories found in each phase, while topic emotional evolution was visualized. RESULTS Based on the OCC model, seven types of emotions were categorized to describe emotional distribution during the Wuhan lockdown: admiration, hope, joy, neutral, fear, reproach, and distress. Among these, admiration and reproach held the largest proportions of emotional expression. Further, expressions of emotion were significantly related to users’ gender, location, and whether or not their account was verified. The lockdown was divided into five phases, incubation phase, explosive phase, declining phase, stable phase, and unblocked phase, which showed citizens emotions transition from reproach and fear, through reproach and admiration, hope and admiration, to reproach and admiration, then to joy and admiration. Admiration and joy increased while fear declined. There were statistically significant correlations between different emotions within the subtle emotional categories. The topics showed that public attitudes towards the lockdown gradually improved. In addition, emotional evolution was influenced by topics, but not limited to them. CONCLUSIONS This study revealed insights into public emotions expressed on Sina Weibo during the lockdown of Wuhan in January 2020. Seven emotion categories were determined, providing governments with greater appreciation of citizen emotions to support them and develop appropriate policies to minimize stress and anxiety. The responses of the government of Wuhan were found to comfort citizens during lockdown which can be used as best practice or a case for other countries and regions affected by COVID-19 outbreaks. In addition, emergency agencies should pay continuous attention to citizens lives and psychological status post-pandemic.