Research On The Spatiotemporal Evolution Of Earthquake Public Opinion And Public Information Guidance Based On Emotional Analysis

Earthquake disaster is one of the most serious natural disasters in China,which seriously threatens the life and property safety of our people.In recent years,earthquakes have occurred frequently in China,with a significant increase in the number of strong earthquakes greater than6.0.With the rapid development of social media platforms,the impact of online public opinion on political order and social stability is increasing day by day,becoming one of the important indicators for analyzing hot social issues.Social media data often contain rich information,including subjective emotions,vocabulary features,time scales,spatial scales,etc.After the earthquake disaster,the public will show specific public opinion characteristics on the social media platform.The analysis of public opinion characteristics and spatial distribution after the earthquake will help to understand the emotional response of the people in the disaster area and the development law of public opinion.This article is based on Sina Weibo data,taking examples of the Changning earthquake in Sichuan,the Luxian earthquake in Sichuan,and the 6.4 magnitude earthquake in Yunnan.By extracting and mining information from Weibo text,release time,and geographical location within 24 hours after the earthquake,we conducted blog feature analysis,spatiotemporal distribution analysis,sentiment analysis,and kernel density analysis on the three earthquakes.And in sentiment analysis,an improved text sentiment analysis method based on sentiment dictionaries has been improved.In response to the specific public opinion environment of earthquakes,adjustments have been made to sentiment assignment,improving the accuracy of sentiment analysis and making it more in line with the earthquake public opinion analysis standards.Research has found that by comparing and analyzing the earthquake public opinion of three earthquakes,the conclusions are as follows: First,after the earthquake occurred,the geographic location information generated showed an upward trend over time,and the spatial distribution showed a multipolar differentiation phenomenon.The densely populated areas were the areas that were severely affected by the earthquake or the economic and political centers within the province.Second,compared to the Changning earthquake in 2019 and the Luxian earthquake in2021,the Luxian earthquake showed a growth trend with a lower initial number but a larger increase compared to the Changning earthquake.This is because the earthquake in Luxian County occurred in the early morning,which is an inactive period for users.Third,comparing the Sichuan Luxian earthquake and the Yunnan Yangbi earthquake,which also occurred in 2021 and had different geographical locations,it was found that the Luxian earthquake generated much more data than the Yangbi earthquake.This is because Sichuan Province has better economic development and a higher population compared to Yunnan Province.The user activity in the Sichuan Chongqing region(collectively known as Sichuan and Chongqing)is relatively high,while the user activity in Yunnan province is relatively low.Indicating that the earthquake public opinion situation is closely related to user activity.In terms of Weibo information dissemination,the following characteristics are presented: First,high attention to hot search topics.Second,the downward trend of the original Weibo is obvious.Third,cross-impact of hot issues and public opinion.Based on the above research content,this article proposes countermeasures and suggestions for the different stages of public opinion development and public opinion guidance.The three stages of public opinion development are the sudden period,the extension period,and the dissipation period,corresponding to different response strategies.In terms of public opinion guidance,mainstream media should: First,maintain a professional attitude.Second,provide positive guidance.Third,strengthen publicity and education.This article focuses on earthquake public opinion.Based on Sina Weibo data,it combines geographic information data with text data to improve sentiment analysis methods.Blog feature analysis,sentiment analysis,spatiotemporal analysis,and kernel density analysis were performed on three earthquakes,and the analysis results were visualized.Based on the theoretical knowledge of public information dissemination,this study analyzed the public opinion characteristics and the causes of the three earthquakes and proposed corresponding public opinion guidance suggestions.This helps to enrich the experience in earthquake public opinion response and provides a scientific basis for the government to respond and guide public opinion work.