基于话题特征的网络舆情参与者情感演化分析.PDFVIP

第34卷摇 第11期 情摇 报摇 杂摇 志 Vol.34摇 No.11 2015年11月 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 JOURNAL OF INTELLIGENCE摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 Nov. 摇 2015 基于话题特征词的网络舆情参与者 情感演化分析* 黄卫东摇 林摇 萍摇 董摇 怡摇 李宏伟 (南京邮电大学管理学院摇 南京摇 210023) 摘摇 要摇 网络舆情参与者对舆情事件情感指向的清晰定位,可以为舆情事件主体或监管部门准确掌握民众关注焦 点、制定有效的网络舆情引导措施提供科学的参考。 使用PLSA模型提取网络舆情话题及其特征词,结合TF-IDF 修正话题特征词;基于已经提取到的话题文本特征词,构建对应的情感词表,并应用HowNet相似度算法计算每个情 感词对应的正负情感倾向值;综合考虑特征词对应的所有情感词,计算得出特征词的情感值,对舆情参与者的情感 指向及变化进行准确定位。 关键词摇 网络舆情摇 话题摇 特征词摇 情感分析 中图分类号摇 C931.6摇 摇 摇 摇 摇 摇 文献标识码摇 A摇 摇 摇 摇 摇 摇 文章编号摇 1002-1965(2015)11-0117-06 DOI摇 10.3969/j.issn.1002-1965.2015.11.022 Analysis on the Feature Words Based Evolution of NetizensSentiments in Network Public Topics Huang Weidong摇 Lin Ping摇 Dong Yi摇 Li Hongwei (Management Department of Nanjing University of Postsand Telecommunications,Nanjing摇 210023) Abstract摇 The research on the evolutionofnetizenssentimentsprovidedtheoreticalbasisforrelevantorganizationstounderstandtheneti鄄 zenssentimentsof the topicsthey werefocusing on andtomakeeffective opinion guidance ofthenetwork publictopics. Thispaper used the PLSA model to extract the topicsand thetext-wordsmatrixesafter segmenting thetexts. Combining with the TF-IDF model and the PLSA model,the paper accurately identified the sub topicsand the feature words. Based on the text-wordsmatrixeswhich hadbeen ex鄄 tracted before,thispaper constructedthesentimentvocabularies. TheapplicationofsimilarityalgorithmHowNethelpedustocalculatethe sentiment value of each word from the semanticangle,beingpositive or negative. Takingall ofthefeaturewordsinto consideration,this paper calculated the netizenssentimentsof each feature word in the network public topics. Key wo