摘要
在复杂的网络舆论生态中,突发事件中的舆情发展更具多变性和难以预测性,通用情感词典已难以适应当前突发事件文本情感分析的需要,建立面向突发事件应急管理的专业情感词典,对于提升网民情感分析的准确度和及时把握舆情走向具有重要意义。据此,采用机器采集加人工构建的方式,以近5年的10起暴雨洪涝灾害的微博评论文本为语料,建立“突发事件·暴雨洪涝”情感词典。经检验发现,该词典显著提高了暴雨洪涝文本情感分析的正确率和召回率,为突发事件的应急管理提供了更为精确和可操作的决策基础。
In the complex network public-opinion ecology,the development of public opinion in emergencies is more and more changeable and unpredictable.The general sentiment lexicon has a great difficulty to adapt to the current emotional analysis in unexpected events.It is of great significance to establish an emergency sentiment lexicon,because it can improve the accuracy of sentiment analysis and forecast the public opinion timely.In this paper,we use the method of machine acquisition and artificial construction to establish the“emergency:rainstorm and flood”sentiment lexicon,with 10 rainstorm and flood disasters’microblogging commentary in recent 5 years.It has been found that the emergency sentiment lexicon has significantly improved the correct rate and recall rate of the emotional analysis of the rainstorm and flooding disasters’microblogging commentary,and provided a more accurate and operational basis for the emergency management.
作者
周莉
杨小俪
ZHOU Li;YANG Xiao-li(School of Journalism and Communication,Central China Normal University,Wuhan 430079,Hubei,China)
出处
《武汉理工大学学报（社会科学版）》
2019年第4期8-14,共7页
Journal of Wuhan University of Technology:Social Sciences Edition
基金
国家社会科学基金一般项目“社会动员中的网络情绪研究”(16BXW078)
关键词
突发事件
应急管理
情感词典
暴雨洪涝
网络舆情
emergency
emergency management
sentiment lexicon
rainstorm and flooding disasters
internet public opinion
