突发公共事件多情景下的网络舆情危机预警研究  

Research on the Early Warning of Network Public Opinion Crisis under Multiple Scenarios of Public Emergencies

作　　者：田世海 于越 邓舒予 Tian Shihai

机构地区：[1]哈尔滨理工大学经济与管理学院,黑龙江哈尔滨150080

出　　处：《情报理论与实践》2023年第1期132-140,共9页Information Studies:Theory & Application

基　　金：教育部人文社会科学研究规划基金项目“重大突发公共卫生事件舆情风险预警机制研究”(项目编号:22YJA630080);黑龙江省自然科学基金项目“融媒体时代突发事件网络舆情引导机制研究”(项目编号:LH2019G017);国家自然科学基金项目“高技术虚拟产业集群运行模式研究”(项目编号:70873029)的成果。

摘　　要：[目的/意义]突发公共事件已成为社会治理难题,探究该事件网络舆情危机预警过程,有助于相关部门科学认知舆情发展态势,提高危机处理能力。[方法/过程]文章首先基于知识元模型,从情景事件、情景驱动要素、情景推理要素、情景目标要素四个方面对网络舆情危机预警情景进行结构性表达,并构建预警情景动态贝叶斯网络模型;其次,结合EM算法对模型进行训练和测试,明确各情景下舆情危机预警重点关注的影响要素;最后,以南京市禄口机场疫情事件为例,对各情景舆情危机预警等级进行评估分析。[结果/结论]训练出来的动态贝叶斯网络具有一定的可靠性,不同情景状态中采取不同的应对方案会对舆情危机预警级别产生不同的影响,为政府及相关部门动态制定危机应急决策提供参考。[Purpose/significance]Public emergencies have become a difficult problem in social governance,and exploring the early warning process of the network public opinion crisis of the incident will help relevant departments to scientifically understand the development trend of public opinion and improve the ability to deal with crises.[Method/process]Based on the knowledge metamodel,this paper first expresses the network public opinion crisis early warning scenario from four aspects:scenario events,scenario driving elements,scenario reasoning elements,and scenario target elements,and constructs a dynamic Bayesian network model of early warning scenarios.Secondly,the model is trained and tested in combination with the EM algorithm,and the influencing factors of public opinion crisis early warning under each scenario are clarified.Finally,taking the epidemic incident at Nanjing Lukou Airport as an example,the early warning level of public opinion crisis in each scenario is evaluated and analyzed.[Result/conclusion]The dynamic Bayesian network trained in this paper has certain reliability,and different response plans in different scenario states will have different impacts on the warning level of public opinion crisis,providing a reference for the government and relevant departments to dynamically make crisis emergency decisions.

关 键 词：突发事件 网络舆情 危机预警 动态贝叶斯网络 知识元 情景推演 

分 类 号：C912.63[经济管理] G206[社会学]