Network Analysis of Emergency Information Dissemination Considering the Strength Relationship Between Nodes postprint

Abstract: Purpose/Significance With the rapid development of new media technology, social media platform has become the main carrier of information dissemination. Social network analysis (SNA) is used to study the information dissemination structure and mode of emergencies in microblog, which provides theoretical support for the government to effectively deal with emergencies and crises. Methods/Process Taking Tangshan barbecue restaurant beating incident as an example, Weibo data were collected to build an information dissemination network with strength relationship between nodes. Social network analysis has been used to analyze the user attributes, node attributes, network attributes and dissemination attributes of the information dissemination network, in order to explore the role of the strength relationship between nodes in emergency information transmission. Results/Conclusions 1) The factors of user gender, activity, and region affect their dissemination power. In particular, female users, users with high activity or influence and those in developed provinces have a stronger power of information transmission. 2) The core nodes that plays the role of bridge in the dissemination chain is particularly critical. Nodes with strong relationships generally occupy central positions in the information dissemination network and may mainly consist of opinion leaders and mainstream media with greater influence. The pathways of weak and authority relationships are mainly concentrated around a few core nodes, while the pathways of strong relationships are dispersed. 3) Emergency information transmission network has high efficiency and sparse characteristics. 4) The whole process of information transmission is still dominated by weak relationships. Authority relations play an important role in all stages of information transmission, while the role of strong relationships is mainly concentrated in the initial stage. The results of this paper help to deepen the understanding of the patterns and rules of emergency information dissemination, and provide some insights for more effective management and guidance of emergency information dissemination in a particular field such as agriculture. However, our research still has shortcomings, such as insufficient crawling of user attributes and insufficient research methods. In the future study, we will obtain more abundant characteristics of users involved in the dissemination such as age, occupation and education level through interviews or questionnaires, and introduce new methods such as machine learning and graph neural network to predict and analyze the transmission path and node relationship of emergency information, so as to cover these shortcomings, improve the comprehensiveness and effectiveness of the study and deeply explore the information dissemination rules of emergencies.