微博环境下新冠肺炎疫情事件对网民情绪的影响分析  

Research on the Influence of COVID-19 Event on the Netizen Emotion under the Microblog Environment

作　　者：刘忠宝 秦权[3] 赵文娟 Liu Zhongbao;Qin Quan;Zhao Wenjuan(Key Laboratory of Cloud Computing and Internet-of-Things Technology(Quanzhou University of Information Engineering),Fujian Province University,Quanzhou 362000;Institute of Language Intelligence,Beijing Language and Culture University,Beijing 100083;School of Software,North University of China,Taiyuan 030051)

机构地区：[1]云计算与物联网技术福建省高等学校重点实验室(泉州信息工程学院),泉州362000 [2]北京语言大学语言智能研究院,北京100083 [3]中北大学软件学院,太原030051

出　　处：《情报杂志》2021年第2期138-145,共8页Journal of Intelligence

基　　金：国家社会科学基金一般项目“大数据环境下面向图书馆资源的跨媒体知识服务研究”(编号:19BTQ012)。

摘　　要：[目的/意义]微博作为一种重要的信息传播载体,在疫情信息发布与传播中发挥着重要作用。深入分析疫情信息中蕴含的疫情事件及其对网民情绪的影响,有助于各级政府准确掌握网络舆论情况,科学高效地做好防控宣传和舆情引导工作。[方法/过程]以新冠肺炎疫情相关的微博新闻及其评论作为研究对象,利用条件随机场(Conditional Random Field,CRF)模型从微博新闻中抽取疫情事件并建立疫情事件画像;在情感词典的基础上,引入双向长短期记忆网络(Bidirectional Long Short-Term Memory,Bi-LSTM)模型建立网民情绪画像;利用基于自注意力机制的Bi-LSTM模型对疫情事件与网民情绪进行关联分析。[结果/结论]真实语料集上的实验结果表明,围绕捐资、防控、临床和英雄等主题,CRF模型疫情事件抽取的F值均达到73%以上,Bi-LSTM模型网民情绪识别的F值均在70%以上,基于注意力机制的Bi-LSTM模型给出的网民情绪分布基本符合疫情发展态势。[Purpose/Significance]Since the outbreak of COVID-19,microblog,as an important carrier of information transmission,has played an indispensable role in the release and dissemination of epidemic information.The in-depth analysis of the epidemic information and the impact on the netizen emotion contained in COVID-19 event can help governments grasp the public opinions on the Internet as well as do a good work in epidemic prevention and public opinion guidance efficiently.[Method/Process]Taking the microblog news and its comments related to the COVID-19 as the research object,this paper then uses Conditional Random Field(CRF)model to extract event information and establish an event portrait for the COVID-19,introduces the Bidirectional Long Short-Term Memory(Bi-LSTM)model to build netizen emotional portrait based on sentiment dictionary,and finally analyzes the relationship between epidemic events and netizen emotions by using an attention-based Bi-LSTM model.[Result/Conclusion]The experimental results on the real corpus with the topics of material,prevention and control,clinical and hero show that the F value of the CRF and Bi-LSTM model has reached more than 73% and 70% respectively.The distribution of netizen emotions is basically in line with the epidemic development based on the attention-based Bi-LSTM model.

关 键 词：微博 新冠肺炎 疫情事件 网民情绪 关联分析 

分 类 号：G353.1[文化科学—情报学]