摘要
为有效识别突发事件信息传播在不同阶段中的关键节点及其演化特征,本文结合危机传播的生命周期,提出一种考虑用户行为特征、网络全局信息以及影响力衰退机制的关键节点动态识别方法。以"哈维"飓风事件为案例进行研究,利用Spearman相关分析和SIR传播模型检验了方法的合理性,并在不同演化阶段关键节点特征对比分析的基础上,提出针对不同阶段突发事件信息传播的舆情治理策略。实验结果表明:与PageRank方法相比,该方法识别出的关键节点在传播速度和传播范围方面表现出一定的优势;随着信息传播不同阶段的演化,关键节点的认证率呈现上升趋势,信息优势表现为先下降后上升,响应优势呈现出相反趋势,而结构优势差异并不显著;在突发事件舆情治理方面,可重点识别潜伏期中高原创、高信息优势和非认证的关键节点,注意搜集爆发期中普通型关键节点掌握的信息,强化蔓延期中各类型关键节点之间的协同,留意消散期中小范围群体的聚集现象。
In order to effectively identify the key nodes in information propagation networks during emergencies and their dynamic characteristics during different stages of an emergency, this paper proposes a method that introduces crisis lifecycle theory and considers characteristics of user behaviors and global network attributes in the information propagation of social networks, as well as the decay law of spreading influence. Hurricane Harvey was chosen as a study case to conduct the experiment. Spearman's correlation analysis and the SIR model were used to verify the effectiveness of this method. The results show that the TPR method is somewhat better than PageRank in terms of spreading speed and spreading scope. With the evolution of different stages of information propagation, the verification rate of key nodes increases. Therefore, the information advantage decreases at first, then increases after the chronic period, while the response advantage shows an opposite trend. There are no significant differences in the aspect of structural advantage. The results shed light on the management of public opinion: administrators should a) focus on the key nodes in the prodromal period that are non-verified and outstanding in terms of originality and information advantage;b) pay more attention to information provided by key nodes that are common netizens in the breakout period;c) strengthen the coordination among different types of key nodes in the chronic period;and d) keep an eye on small-scale clusters during the recovery period
作者
陈思菁
李纲
毛进
巴志超
Chen Sijing;Li Gang;Mao Jin;Ba Zhichao(Center for Studies of Information Resources, Wuhan University, Wuhan 430072)
出处
《情报学报》
CSSCI
CSCD
北大核心
2019年第2期178-190,共13页
Journal of the China Society for Scientific and Technical Information
基金
国家自然科学基金重大课题"国家安全大数据综合信息集成与分析方法"(71790612)
国家自然科学基金青年项目"突发公共卫生事件社交媒体信息主题演化与影响力建模"(71603189)
