Research On The Information Diffusion And Governance Of Online Social Networks Under The Background Of Public Emergencies

In the environment of information society,a trend is formed during the development and evolution of public emergencies:a social network is taken as the carrier,emergencies are the core,and expressions of participants' emotions,attitudes,opinions,and opinions are taken as the forms of expression.The proliferation of information,such as immediacy and interactivity,has changed the environment of public opinion and in turn has influenced the trend of events.Because of these characteristics,online social networks have a profound impact on the response,handling,and guidance of public emergencies.Information governance of online social networks has become an important research content and research hotspot in the field of emergency management.The nature of online social networks is to transplant and enlarge real-world information diffusion patterns in social networks.In this process,information is rapidly iterated and reacts to the real world.Due to the influence of social networks,the information diffusion model of public emergencies has undergone fundamental changes.The traditional research theories and methods cannot be adapted to the development of the new situation.The existing literature on the use of social networks in public emergencies is mostly focused on describing and characterizing the social network's response to public emergencies.It has not yet explored the reasons for the proliferation of information in critical public events,the key influencing factors,and the influence and countermeasures of social networks in emergency management.This study reexamines the process of information diffusion in online social networks in the context of public emergencies from the perspective of diffusion driving force and proposes governance strategies based on research findings.First of all,theoretical hypotheses are proposed for users' influence and users'information diffusion behavior attributes that impact on the motivation of information diffusion on social networks,and hypotheses are then verified through empirical research.The results show that the speed of information dissemination depends on the influence of the information publisher.When information is published by more influential users in the network,it spreads faster.In addition,the information released in the early stages of an emergency is more rapid than the information released in the middle and later stages.Secondly,the geographic location of the information publisher has a significant impact on the rate of information diffusion.The dissemination of information released by publishers inside the region affected by an emergency is faster than information published outside the region.Secondly,aiming at the lack of research on the interaction between user's emotional evolution and information diffusion,a model combining emotional evolution and information diffusion process is proposed to reveal the impact of the evolution of user sentiment on social network information during emergencies.Taking Sina Weibo and Twitter as research objects in the study,through simulation verification of the information diffusion process in social networks,the impact of different emotional distributions and emotional interventions on information diffusion was explored.The research results show that users' information with basically the same expression of emotion spreads faster and wider.Emotional intervention can effectively guide information diffusion,and the earlier the intervention,the better the guiding effect.In addition,aiming at the problem of how the user's privacy awareness influences the information diffusion in social networks,an information diffusion model that incorporates user privacy awareness is proposed.The study first calculates privacy scores based on user attributes,and then measures the likelihood of their participation in network proliferation.Then four kinds of simulation networks with different structure and degree distribution were constructed.Through simulation experiments,the influence of user's privacy attitude to information diffusion was studied.The results of the study show that privacy attitudes can have an impact on the diffusion of information.Higher privacy awareness among users in the network can lead to a reduction in the efficiency of information diffusion and vice versa.This conclusion applies to all four simulation networks studied,but it is most evident in the microblog network.Then,for the problem that the information diffusion model is not efficient and accurate,this paper proposes a method based on Bayesian networks for calculating the information diffusion probability of different contents in social networks.The method uses the text content of the information,the spread history of the network,and the characteristics of the network and users to construct a Bayesian network model and uses the hidden information of the text content to calculate various features such as user similarity,content similarity,user interest,and the like.The study conducted an experimental assessment of the method using the microblogging dataset of the emergencies.The experimental results show that the proposed method performs well for the data sets of various event categories and is more efficient and accurate than the existing methods.It shows that the method is effective in describing the factors of information diffusion in the context of emergencies and can better predict the proliferation of information because it takes into account the diffuse information content,user interest,and network dynamics-related features.Then,users' characteristics that affect information diffusion are summarized based on the above research results,a Bayesian network-based modeling method is proposed,and an information diffusion model for social networks in the context of emergencies is built in this paper.The text content of information,the history of network diffusion as well as the characteristics of networks and users are used in the construction of the Bayesian network model,and the hidden information of the text content is used to calculate various characteristics such as user similarity,content similarity and users'interest.On top of that,the data sets of emergencies from Sina Weibo are used to evaluate the model,and the results show that for the data sets of different types of events,the proposed model is better than existing ones in terms of efficiency and accuracy,which means the method is effective in describing the factors of information diffusion in the context of emergencies,takes into account the users' characteristics,information content,users' interest and network dynamics-related characteristics of diffusion,and can better make a prediction for information diffusion.Finally,based on the above research conclusions,the key factors and subdividing factors affecting information diffusion are summarized and analyzed,and the governance strategies for online social networks in the context of public emergencies are proposed.This study provides theoretical support and decision support in intervention,guidance,and prediction of emergencies for the information diffusion of online social networks,and provides new ideas and methods for the monitoring,management,and early warning of public emergencies.