Policy Perception During Global Health Emergencies: Text Analysis Based on Weibo Data

The rise of social media platforms has provided a new channel for interaction between the government and the public. By analyzing public reactions on social media through natural language processing technology, scientific evidence can be provided for the formulation of government policies. This paper is based on 126,454 text data posted on Sina Weibo by mainstream official media from January to December 2020. It measures policy perception from the aspects of extent and direction and explores the differences in the impact of various policy perceptions on the stock market, making policy perception a quantifiable and analyzable data sample. The results show that positive information released by official media has a higher degree of perception, indicating that positive signals about the pandemic are more easily disseminated and accepted by the public. This study aims to provide strategic recommendations for the government in formulating policies, stabilizing public emotions, and promoting the stable development of the financial market in response to major public health events.