Characterizing Chinese online public opinions towards the COVID-19 recovery policy

Abstract

Purpose The online users’ characteristic information can provide decision support for policy-designing and construction of public strategies. Hence, this paper aims to conduct online public opinion mining on the recovery policy stimulating the economies stroked by COVID-19 epidemic. Also, sentimental analysis is performed to uncover the posters’ emotion towards the target policy. Design/methodology/approach This paper adopts bidirectional encoder representations from transformers (BERT) as classifier in classification tasks, including misinformation detection, subject analysis and sentimental analysis. Meanwhile, latent Dirichlet allocation method and sentiment formulations are implemented in topic modelling and sentiment analysis. Findings The experimental results indicate that public opinion is mainly non-negative to the target policy. The positive emotions mainly focus on the benefits that the recovery policy might bring to stimulate economy. On the other hand, some negative opinions concerned about the shortcomings and inconvenience of the target policy. Originality/value The authors figured out the key factors focused by the public opinion on the target recovery policy. Also, the authors indicated pros and cons of the recovery policy by analysing the emotion and the corresponding topics of the public opinion on social media. The findings of the paper can be generalized in other countries theoretically to help them design recovery policy against COVID-19.

Characteristics of Collective Resilience and Its Influencing Factors from the Perspective of Psychological Emotion: A Case Study of COVID-19 in China

“…Since the outbreak of COVID-19, the extraction of the semantic or emotional content of textual commentaries posted by the public has become a useful research tool to explore the evolution of topics and the emotional expression characteristics of particular groups Due to the prolonged series of COVID-19 lockdowns, social media has also become a placeholder for collective resilient processes modulated by cognitive and emotional components [ 42 ]. This article also draws on a similar research approach, using Weibo data and Baidu indexes, to investigate the collective resilience characteristics of social groups during an emergent public crisis, taking randomly-selected unstructured groups with different social attributesfrom the Weibo platform as research subjects [ 25 , 43 , 44 , 45 , 46 , 47 , 48 , 49 ].…”

Constructing a spatiotemporal situational awareness framework to sense the dynamic evolution of online public opinion on social media

“…Online public opinion is a collection of public sentiments, attitudes and opinions about social events expressed through internet platforms, which can help public organisations and authorities to understand public perceptions of social events and thus assist in decision-making (Li et al , 2020; Yu et al , 2022). With the rapid development of the internet, social media has become a major gathering place of public opinion.…”