突发公共卫生事件中多主体参与的社交网络群体聚类模型

A Social Network Group Clustering Model for Multi-subject Participation in Public Health Emergencies

[目的/意义]通过用户聚类,调查不同类型的参与主体、不同参与时间以及不同影响力的主体在突发公共卫生事件信息传播中的作用。[方法/过程]文章将参与新型冠状病毒肺炎疫情(COVID-19)的主体划分为4种类型,分别是政府官微(GO)、主流媒体(MM)、意见领袖(OL)和普通网民(ON),将主体参与时间、影响力和类型作为聚类系数,通过Canopy和K-means聚类算法进行用户群体聚类。[结果/结论]研究发现,在紧急突发事件期间,政府官微(GO)、主流媒体(MM)和意见领袖(OL)是事件信息源,在信息传播网络中占主导地位,其中政府的主导作用最强。公众则主要助力于信息的转发和扩散,并且其信息来源于其他三类主体。对于信息传播主体而言,越早发布信息就越容易被公众接受,进而拥有更强的影响力。

[Purpose/significance]Through user clustering,the role of different types of participants,different times of participation,and different influential participants in the dissemination of information on public health emergencies was investigated.[Method/process]In this study,the participants participating in COVID-19 are divided into four categories,namely government officials(GO),mainstream media(MM),opinion leader(OL)and ordinary netizens(ON).Then,the participation time,influence and category are used as clustering coefficients,and user clustering is performed through Canopy and K-means clustering algorithms.[Result/conclusion]The study found that during emergencies,government officials(GO),mainstream media(MM)and opinion leader(OL)are the sources of incident information.And they dominate the information dissemination network,of which the government has the strongest leading role.For the participants of information dissemination,the sooner the information is released,the easier it is to be accepted by the public,and thus to influence more people.