主管:中国科学院
主办:中国优选法统筹法与经济数学研究会
   中国科学院科技战略咨询研究院

中国管理科学 ›› 2023, Vol. 31 ›› Issue (11): 58-66.doi: 10.16381/j.cnki.issn1003-207x.2021.2045

• • 上一篇    下一篇

重大突发事件舆情风险点预测研究

马宁1,刘怡君1,2(),李凉凉3   

  1. 1.中国科学院科技战略咨询研究院, 北京 100190
    2.中国科学院大学公共政策与管理学院, 北京 100049
    3.中国人民大学商学院, 北京 100872
  • 收稿日期:2021-10-09 修回日期:2021-11-11 出版日期:2023-11-15 发布日期:2023-12-05
  • 通讯作者: 刘怡君 E-mail:yijunliu@casisd.cn
  • 基金资助:
    国家自然科学基金资助面上项目(72074206);国家自然科学基金资助重大项目(T2293772)

Research on the Risk Points Prediction of Emergency Public Opinion

Ning MA1,Yi-jun LIU1,2(),Liang-liang LI3   

  1. 1.Institute of Policy and Management, Chinese Academy of Sciences, Beijing 100190, China
    2.School of Public Policy and Management, University of Chinese Academy of Sciences, Beijing 100049, China
    3.School of Business, Renmin University of China, Beijing 100872, China
  • Received:2021-10-09 Revised:2021-11-11 Online:2023-11-15 Published:2023-12-05
  • Contact: Yi-jun LIU E-mail:yijunliu@casisd.cn

摘要:

重大突发事件发生后,其舆情传播中呈现多风险点共存现象,公众的非理性情绪被不断放大,给舆论生态造成消极影响。如何在突发事件发生后的第一时间,准确预测其可能衍生的舆情风险点,成为决策者有的放矢、高效应对的关键。本文以我国积累10余年的历史突发事件大数据为基础,识别出突发事件舆情传播中的各类风险点,并对风险点做共现分析;然后应用特征相似度算法,计算新发生突发事件与历史突发事件之间的相似度,并结合风险点共现率,定量预测突发事件中所有的显性和隐性的舆情风险点。本文“以史为鉴”,从舆情风险源头治理的视角,期望在潜在风险萌芽期预测舆情风险点,为掌握舆情风险处置工作的主动权和话语权提供帮助。

关键词: 突发事件, 风险点, Jaccard共现指数, 特征相似度算法, 预测

Abstract:

After the occurrence of emergencies, coexistence of multiple risk points appears in public opinion communication, which amplifies irrational emotions of the public and causes negative impacts on the ecology of public opinions. In this context, how to accurately predict possible public opinion risk points derived from emergencies in the first time after the occurrence of emergencies has become a targeted and efficient key. In this thesis, based on historical data of emergencies that took place in recent ten years in China, various risk points in the public opinion communication of emergencies are identified and a co-occurrence analysis on risk points is conducted. Secondly, feature similarity algorithm is utilized to calculate the similarity between different emergencies, while Jaccard index is used to quantitatively predict all the explicit and potential public opinion risk points in emergencies. By taking history as a mirror, this research aims to predict public opinion risk points in the budding stage of risks from the perspective of source governance of public opinion risks, with the hope of offering help for grasping the initiative and the right to speak when coping with public opinion risks.

Key words: emergencies, risk points, jaccard index, feature Similarity algorithm, prediction

中图分类号: