突发事件中政府信息发布对社会负面情绪感染-演化的影响机制研究

The breeding and spreading of negative emotion in public emergency events posed severe challenges to social governance. The traditional government information release strategies ignored the negative emotion evolution mechanism,and having the following drawbacks：(1)the current regulation strategies are more descriptive, speculative and less operational; (2) there exists gap between information transition and emotion spreading; (3) there is short of accuracy in research; and (4) the releasing strategies are quite lagging behind, and etc. From the perspective of news framing effects on the negative emotion regulation, the research takes government information releasing policy as studying subject, and applies SVM classifier, deep learning method into news framing framework construction process, and tries to establish three-layers news framing and finally achieves the combination of discourse qualitative and framing quantitative study. Based on the psychology experiment, the research will test the regulation effect of government information release on negative emotion's contagion and will lay a psychological foundation for the simulation model. This project can also enrich the Aging theory, set up the cycle life modeling and extract the features of negative emotion in public emergency events. By using the super-network, multi-agent modeling, this research tends to set up a simulation modeling, to analyze the intervention influences of government information release on negative emotion's evolution, to organize and train release rule sets, and finally to form reasonable, measurable government information release strategies.. The research result will provide theoretical implications and technical supports for the social governance. And it can also help to establish negative emotion management mode, and construct a new pattern of the public opinion guidance.

突发事件中负面情绪的滋生、蔓延，对创新社会治理提出了严峻挑战。因忽视其对社会负面情绪感染-演化的影响机理，现有信息发布策略多为描述性、思辨性论述，操作性弱，存在发布滞后、精准性弱、公众关切研判有误等缺陷。为此，基于媒体框架对负面情绪的调节视角，将政府信息发布实践中的“实然”、“应然”相结合，研究突发事件中政府信息发布对调节负面情绪感染、干预负面情绪演化的机制。首先，采用深度学习、情感计算等方法，建构政府信息发布的媒体框架体系；设计心理实验，验证政府信息发布对负面情绪感染的调节作用；拓展Aging theory理论，建立演化模型，提炼负面情绪演化的结构特征，解析政府信息发布对负面情绪演化的干预机理；利用超网络、多主体建模方法，构建仿真模型；结合案例研究，开发原型系统，形成科学合理的政府信息发布策略。研究成果将为健全社会治理机制提供理论基础，对于提升突发事件应急管理水平，具有重要实践指导意义。

突发事件的不可预知性和危害性与政府信息发布内容相耦合，易造成负面情绪的滋生、蔓延。针对现有信息发布策略多定性描述，存在时机滞后、精准性弱等问题。因此本项目进行了突发事件中政府信息发布对社会负面情绪感染-演化的影响机制研究。针对词频-文档矩阵、词袋模型等传统方法不能保留词序的信息，且会面临严重的数据稀疏问题的缺陷，采用基于深度学习方法，通过对词语的上下文以及词语与上下文的句法关系考察，使用基于情感词汇本体和深度学习算法，构建情绪词典；针对微博短文本的特征，基于依存句法、方面情感分析等思想，提出了细粒度情绪计算模型、面向微博的方面情感计算模型；基于媒体框架理论，运用心理学实验、内容分析等方法，解析了突发事件下负面偏向的形成机理，检验了不同媒体框架对风险认知的影响以及情绪的中介作用；借鉴传统传染病模型的基础上，采用复杂网络理论，构建了情绪传播模型；在此基础上，以新冠疫情为背景，构建实证模型，分析了不同防疫政策下情绪演化规律及调试作用。在项目研究过程中，积极参加学术会议，进行学术交流。同时，多名博士、硕士研究生参与了项目的研究工作，培养了多名研究生。同时，基于本项目的研究成果，已经发表论文16篇，其中期刊论文11篇（SCIE源期刊4篇，SSCI源期刊5篇，CSSCI源期刊1篇），会议论文5篇。目前还有4篇论文已投稿到国际SCIE/SSCI检索期刊。还有4篇工作论文待投稿。本项目成果对于提升突发事件应急管理水平，完善社会心理服务体系具有重要实践指导意义。