摘要
[目的/意义]网民情感变化是影响政府舆情应对进程、政策、策略的关键。因此,构建科学高效的情感词典,对网民情感分析研究及应对策略的选择具有重要的实际意义。[方法/过程]结合扎根理论的质性研究特点,在情感词典的构造中融入突发事件演化规律影响因素,采用点互信息算法,TF-IDF,统计量等方法对微博表情符号词典及突发事件专属情感词典进行构建,编制了突发事件情感词典,随后选取"6. 22"杭州保姆纵火案微博语料进行情感分析。[结论/结果]实验发现,与不加入影响因素的情感词典相比,本文构造的词典在准确率召回率等指标的对比中均得到了一定程度的提高。同时,结合扎根理论与主题分析的结果,对舆情发展的不同阶段所选择的舆情应对策略提供了参考。
[Purpose/Significance]The emotional change of netizens is the key to the policy of public opinion.Therefore building a scientific and efficient sentiment lexicon will have great practical significance to the analysis of Internet users and the choice of coping strategies.[Method/Process]Combining the qualitative research characteristics of grounded theory,this paper took the factors influencing the evolution of emergencies into consideration in the construction of sentiment dictionaries.The point mutual information algorithm,TF-IDF,statistics and other methods were used to construct the micro-blog Emotions dictionary and the special sentiment lexicon of emergency,thus this paper compiled the Emotion Dictionary of the emergency.Then the“6.22 Hangzhou nanny arson case”was selected for emotional analysis.[Result/Conclusion]It was found that the sentiment lexicon constructed in this paper had a certain degree of improvement in comparison with the index of accuracy and recall than the previous dictionary.At the same time,combined with the results of root theory and topic extraction analysis,the lexicon would provide reference for the coping strategies chosen in different stages of public opinion development.
作者
张鹏
崔彦琛
兰月新
吴立志
Zhang Peng;Cui Yanchen;Lan Yuexin;Wu Lizhi(Chinese People s Police University,Langfang 065000,China)
出处
《现代情报》
CSSCI
2019年第3期122-131,143,共11页
Journal of Modern Information
基金
教育部人文社会科学基金"面向突发事件的网络流言风险预警及对策研究"(项目编号:17YJC630214)
公安部理论与软科学项目"基于网络舆情综合研判的群体性事件预警研究"(项目编号:2015LLYJWJXY034)
全国统计科学研究重点项目"舆情大数据环境下突发事件民意监测与评估研究"(项目编号:2017LZ37)
河北省科技计划项目"新媒体环境下突发事件危机信息挖掘与决策关键技术研究"(项目编号:17455610)
武警学院国家社会科学基金培育项目"大数据环境下公共危机事件情报挖掘与决策预警研究"(项目编号:SKJJPY201721)
关键词
扎根理论
情感词典
情感分析
微博
突发事件
grounded theory
sentiment lexicon
sentiment analysis
Micro-blog
emergency
