The Coronavirus Disease (COVID-19) is an urgent global public health crisis.
During the outbreak, China actively organized anti-epidemic activities,
and gradually achieved the staged victory of epidemic prevention and control.
It is of practical significance to objectively understand the
public opinion response and regional differences for
improving policy control and scientific governance
during major public health events.
In this study, originating in microblogs data,
a topic extraction and classification model is constructed based on
the Latent Dirichlet Allocation (LDA) topic model and the random forest algorithm.
The hierarchical processing method from total to sub
is adopted to identify 7 primary and 12 secondary theme
topics about public opinion in microblog.
The general distribution of public opinion is analyzed
in terms of amount, space, time sequence,
and content from January 9 to March 10, 2020,
and its regional distribution characteristics were explored
in key regions like Hubei Province,
four major urban agglomerations and border ports.
The results show that the change of microblogs counts
on each topic is positively related to the evolution
of the COVID-19 epidemic, and the temporal and spatial
distribution of public opinion is related to the severity
of the epidemic, the degree of population aggregation,
and the level of economic development.
The spatial distribution of all topics is significantly
consistent, but the spatial distribution within these
areas is obviously different. The response of Chinese
people is rational and positive. It is also found that
the government response lags behind the social media,
the imbalance of resource allocation caused by the sharp
rise of relief information in the short term is prominent,
and the difference in response policies of various urban
agglomeration areas combined with its own regional
characteristics are not obvious.
It is suggested to continue to strengthen the focus of public
opinion on epidemic situation in key areas and the
differentiated and accurate response to local conditions.
The link of the article is:
Provided by the IKCEST Disaster Risk Reduction Knowledge Service System