Accepted for/Published in: Journal of Medical Internet Research
Date Submitted: Mar 23, 2023
Date Accepted: Dec 20, 2023
Social Media Public Opinion about COVID-19 in China on Microblog Platform: Topic Modeling and Multidimensional Sentiment Analysis
ABSTRACT
Background:
The COVID-19 epidemic has attracted wide attention from all over the world, and social media has become an effective medium for the dissemination of public opinion. Sentiment analysis based on social media data can help the government to reasonably control public opinion.
Objective:
This study aims to determine the public opinion topics related to COVID-19 in Microblogs and explore their evolution process from the perspective of time and space, and finally provide reference for epidemic prevention and control.
Methods:
Firstly, a multidimensional analysis model of public opinion in social media was constructed, and 100 000 Microblog comments on COVID-19 topics were collected, and 2000 were screened as sample data. Secondly, the Dalian University of Technology sentiment dictionary was used for public opinion analysis. Then, Latent Dirichlet Allocation (LDA) topic modeling was used to extract feature words. Finally, the ML-LR model fused with sparse matrix was used to explore the evolution trend of social media public opinion, and the high accuracy of the model was verified by indicator calculation.
Results:
The geographical spatial distribution of public opinion accurately mapped the severity of the epidemic in different regions, and at different stages of the epidemic development, the emotional characteristics of the public were quite different, and the overall change was from negative to positive.
Conclusions:
The analysis method of social media public opinion proposed in this paper can effectively reflect the characteristics of public opinion in different regions and different periods and provide theoretical reference for the government to respond to major public safety and health incidents and reasonably control public opinion.
Citation
The author of this paper has made a PDF available, but requires the user to login, or create an account.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.