Abstracts Geography

Spatial and Sentiment Analysis of Public Opinion toward Covid-19 Pandemic (A Study on Ohio and Michigan Twitter Data at the Early Stage of Vaccination)

by Shaghayegh Jabalameli

After almost two years into the pandemic, the whole world still struggles with the outbreak of COVID-19 and its new variants. During this crisis, social media, like Twitter, have been the platforms on which people have been able to share their thoughts and obtain information. This study provides a detailed spatial-temporal analysis of the Twitter online discourse in Ohio and Michigan, (containing approximately 280 thousand tweets), at an early stage of the vaccination rollout, (12th of January 2021, till 10th of March), at the county level. This work aims to explore how people were feeling about the pandemic, the most frequent topics people were talking about, and how the topics were spatially distributed over these two states. Moreover, state government responses and important news during vaccination phases were gathered and analyzed based on the temporal analysis of the tweets. The analysis showed that most of the changes in COVID19-related tweets trends can be explained by related policies implemented by local authorities. Natural language processing using the LDA (Latent Dirichlet Allocation) method for topic extraction and classification model was deployed to identify 11 topics and 8 sub-topics in the extracted Twitter data. According to the temporal analysis of public opinion, the frequency of topics and sub-topics have changed over time due to sensitivity to significant state events and news, and the local government's reactions to the pandemic. Moreover, the spatial distribution of Coronavirus-related tweets and sentiments shows concentrations in the more populated urban areas with a high rate of COVID-19 cases in Ohio and Michigan. People in both states during this period had more positive sentiments than negative sentiments, but the spatial analysis of sentiments shows more dispersion of negative sentiments across Michigan. The government's economic responses to the pandemic, the vaccination timeline phases specified by each state, and the pandemic-related information can contribute to public opinion and sentiment trends. The results of the spatial analysis of public opinions show the spatial distribution similarities between some of the topics and sub-topics. The findings of this study can help understand public demands, opinions, and reactions, follow the impacts of their policies at the county level and their spatial distribution and manage their future responses to the pandemic.