| ObjectiveSocial media platforms represented by microblogs have gradually become the main space for the development and evolution of public opinion on public health emergencies.In order to be able to identify the key players in the process of public opinion dissemination in a timely manner when public opinion occurs,it provides an effective basis for the relevant departments to further develop public opinion management and intervention measures.MethodIn this study,the web crawler technology was used to obtain the comment data and the characteristic data of comment users of 10 popular microblogs during the period from the closure of Wuhan to the announcement of the lifting of the closure,and the sentiment analysis of the comment data was carried out to calculate their sentiment tendency.The calculated sentiment tendency was correlated with user characteristics.The sentiment tendency was used as the weight to construct a user relationship matrix and a network community diagram using social network analysis,and several indicators including centrality,structural holes,density and distance,core and edge were analyzed to identify the key nodes in the network and analyze the composition and overall interaction of the network.Result1.There is a correlation between the emotional tendency of microblog users and six user characteristics: gender,age,number of followers,number of comments and likes,introduction and label.2.A 64×64 user relationship matrix was constructed,and the core nodes of the network community graph of COVID-19 events included CCTV News,People’s Daily,Smart Learning,random drop of cheese fragments,along the way you ZL,Didi Kangaroo,etc.3.The results of point degree centrality showed that the point degree of random falling cheese fragments in the COVID-19 event network had the highest centrality value,followed by nodes such as you ZL and China News network along the way.The nodes with the highest degree of intermediary centrality are the intelligent Learning,People’s Daily,you ZL along the way,China News Network and so on.These nodes play the role of bridge in the communication of public opinion and are in the center of controlling the communication of public opinion.The results of proximity to centrality show that smart students,People’s Daily,China News Network,along the way you have low proximity to centrality value,these points are closely connected with other members of the network,and the less easy to be controlled by others.4.The measurement results of the structural hole show that in the COVID-19 event network,the effective scale value of the smart student user node is 5,and the limit system is 0.211,followed by Didi Kangaroo,China News Network,People’s Daily,CCTV News and other nodes also have the same characteristics,the limit of effective scale value is small,these nodes are not easy to be controlled by others.It is easier to obtain information resources and occupy more structural holes,which strongly proves that they are core nodes.5.The network density of COVID-19 event is 0.0176.The overall connection of the network is relatively loose,and the relationship can be established through the average of two nodes of any two nodes.6.The core edge analysis results show that the core nodes of the COVID-19 epidemic transmission network include People’s Daily,CCTV News,Smart Learning,Didi Kangaroo,a thousand aspects Strategy,J2K＿Xiaodong,Xuejinzi,random drop of cheese fragments,you ZL and so on.Conclusion1.On the whole,participants in public opinion of COVID-19 are diversified,and the transmission network of COVID-19 is sparse and relatively complete.2.The key nodes in the communication network of COVID-19 are at the core of the public opinion communication network,and the role of emotional influence is more significant.3.The governance of public opinion can not only start from the relevant government departments,but also from big V and central influencer in the network.According to the characteristics of users,corresponding measures can also be taken for classification and guidance. |
PDF Full Text Request