新型冠状病毒肺炎疫情期对复工复学微博评论数据的情感分析

Sentiment analysis based on the data of Micro-blog comments on returning to work and school during the COVID-19 epidemic

目的:探究新型冠状病毒肺炎(COVID-19)流行期间复工复学状态下的民众情绪,为心理疏导、宏观调控提供依据。方法:通过Python爬取新浪微博用户在5月26日至6月26日间针对复工复学问题发表的公开评论,运用文本分析的方法计算用户的情感倾向性、情感强度和情绪词频,按照空间、时间维度进行情感可视化分析。结果:研究发现民众对复工复学整体情绪偏向积极,但负性情绪仍不可忽视。根据收集到的文本数据呈现出的情感特征,使用K-均值聚类算法对收集到的31个评论数据地区进行聚类,分成消极、积极以及中性区域,然后运用方差分析并事后比较验证了三类地区文本数据中的负性评论比例存在显著差异(F=18.366,P<0.05)。疫情反弹地区的用户出现负面情绪的指数较高,具有一定的空间、时间维度差异。结论:本研究验证了运用网络数据开展情感分析工作的可行性和有效性,同时表明官方媒体要加强信息平台建设,引导大众理性看待疫情常态化、适应局部反弹情况下经济复苏的社会新形势。

Objective:To explore the public sentiment on returning to work and school during the COVID-19 epidemic,which provides the basis for the government to conduct psychological counseling and macro-control.Methods:Based on the Python text analysis method,Sina Weibo users’data,such as comments on returning to work and school from May 26 to June 26,were collected through Python.The text analysis method was used to calculate the emotional orientation,emotional intensity,and emotional word frequency.Besides,the visual analysis of emotions was carried out according to spatial and temporal dimensions.Results:According to the emotional characteristics of the collected text data,this paper used K-means clustering algorithm to cluster 31 comment data regions,which were divided into negative,positive and neutral regions,and then used ANOVA and post comparison to verify that there were significant differences in the proportion of negative comments among the three regions(F=18.366,P<0.05).The index of negative emotions of users in epidemic rebound areas was higher,with certain spatial and temporal differences.Conclusion:This study verifies the feasibility and effectiveness of using network data for sentiment analysis.At the same time,it find that the construction of information platforms should be strengthened in the official media in order to guide the public to rationally view the normalization of the epidemic situation and to be adapted to the situation of returning to work and school under the local rebound of COVID-19.