| The development of social network has led to the online communication of public opinion,citizens use network platform to communicate with each other,thus form the network public opinion.Due to its multiple social attributes,if not handled properly,it will have negative social impact.the users’ behavior is an important factor in the development of public opinion.It is of positive significance for public opinion supervision and governance to study users’ behavior of public opinion.This paper focus on users’ interaction behavior and discusses the characteristics of public users’ interaction behavior and the influencing model of interaction behavior.in the chapter of users’behavior,by crawling typical public opinion data from the topics——"Henan Rainstorm mutual aid","Shanxi Rainstorm" and"Northeast power restriction",using statistical analysis,social network analysis,text mining and other methods to carry out research,the main conclusions include:the evolution of online public opinion conforms to the law of life cycle;The time distribution of users’posting is fragmented,the nature of public opinion events and the scope of participants have a certain influence on the time distribution of users’ posting;The behavior of "like" is the mainstream interaction behavior,while the behavior of forwarding and comment is different in different types of events.The users are more likely to forward the posts related to disaster relief,information diffusion,material help,and more likely to comment on the posts related to public opinion and sentiment.In public opinion events,only a small number of users get a large number of interactions,while highly interacted users are concentrated in official media and entertainment celebrities with prominent centrality.in the chapter of the influencing model of users’ interaction,theories such as social identity and herd effect are introduced to construct theoretical model to explore the influencing factors of users’interaction behavior in public opinion events.after the construction of theoretical model,we compute the model by using the data collected from 416 valid questionnaire data and amos is used to discover the model,the experiment found that self-efficacy and information quality in network public opinion significantly affect users’information interaction intention,community trust and herd effect have no significant influence on interaction intention,interaction intention significantly affects interaction behavior.Finally,this study integrates the results of the two parts to explain the law of the public users’interaction,the conclusion can give the regulators and operation platform the corresponding theoretical support to optimize the management of public opinion. |
