[目的/意义] 探索网络舆情演化的内在机理能够对舆情发展获得更深层的洞见，有助于相关单位或部门在舆情事件中实现科学的决策与引导。[方法/过程] 研究工作选取特定舆情事件的当事人博文及对应的转发评论，基于情感词典计算评论文本的情感分值。采用时间序列的视角，对舆情发展过程中情感极性的变化以及当事人回应对网民情感的影响进行动态分析。[结果/结论] 研究结果表明，当事人的回应对网民的关注程度有直接影响；舆情发展过程中网民的情感极性并非一成不变；回应的内容直接影响网民的情感，有效的证据和诚恳的态度有助于平抑舆情中的负面消极情绪。
[Purpose/significance] Exploring the internal mechanism of the evolution of Internet public opinion can gain deeper insights into the development of public opinion and help relevant units or departments to achieve scientific decision-making and guidance in public opinion events.[Method/process] This paper selected the parties' blog posts and the corresponding reposted comments of specific public opinion events, and calculated the sentiment score of the review text based on the sentiment dictionary. Using the perspective of time series, dynamic analysis was made on the change of emotional polarity and the impact of the party's responses on the emotions of netizens during the development of public opinion.[Result/conclusion] The results show that the party's responses have a direct impact on the emotional polarity of netizens; the emotional polarity of netizens during the development of public opinion is not static; the contents of the responses directly affect the netizens' emotional polarity, valid evidence and sincere attitudes help calm negative sentiments in public opinion.
[1] 中国共产党第十六届中央委员会第四次全体会议. 中共中央关于加强党的执政能力建设的决定[N]. 人民日报, 2004-09-27(1).
[2] 许鑫, 章成志, 李雯静. 国内网络舆情研究的回顾与展望[J]. 情报理论与实践, 2009, 32(3):115-120.
[3] 田依林, 滕广青. 影响线下衍生行为的网络关键信息分析[J]. 情报理论与实践, 2016, 39(10):80-85.
[4] MULDER K. The dynamics of public opinion on nuclear power. Interpreting an experiment in the Netherlands[J]. Technological forecasting & social change, 2012, 79(8):1513-1524.
[5] KIM K, YOUNG M B, NARAE K. Online news diffusion dynamics and public opinion formation:a case study of the controversy over judges' personal opinion expression on SNS in Korea[J]. The social science journal, 2015, 52(2):205-216.
[6] 刘兵. 情感分析:挖掘观点、情感和情绪[M]. 北京:机械工业出版社, 2017:1-26.
[7] TURNEY P D, LITTMAN M L. Measuring praise and criticism:Inference of semantic orientation from association[J]. ACM Transactions on information systems, 2003, 21(4):315-346.
[8] TABOADA M, BROOKE J, TOFILOSKI M. Lexicon-based methods for sentiment analysis[J]. Computational linguistics, 2011, 37(2):267-307.
[9] 钟敏娟, 万常选, 刘德喜. 基于关联规则挖掘和极性分析的商品评论情感词典构建[J]. 情报学报, 2016, 35(5):501-509.
[10] 张海涛, 刘雅姝, 张枭慧, 等. 基于模块度的话题发现及网民情感波动研究——以新浪微博"中美间贸易摩擦"话题为例[J]. 图书情报工作, 2019, 63(4):6-14.
[11] 张鹏, 崔彦琛, 兰月新, 等. 基于扎根理论与词典构建的微博突发事件情感分析与舆情引导策略[J]. 现代情报, 2019, 39(3):122-131, 143.
[12] 姜金贵, 闫思琦. 基于主题和情绪相互作用的微博舆情演化研究——以"红黄蓝虐童事件"为例[J]. 情报杂志, 2018, 37(12):118-123.
[13] 安璐, 吴林. 融合主题与情感特征的突发事件微博舆情演化分析[J]. 图书情报工作, 2017, 61(15):120-129.
[14] 王仁武, 宋家怡, 陈川宝. 基于Word2vec的情感分析在品牌认知中的应用研究[J]. 图书情报工作, 2017, 61(22):6-12.
[15] MULLEN T, NIGEL C. Sentiment analysis using support vector machines with diverse information sources[C]//Empirical methods in natural language processing. Barcelona:Association for Computational Linguistics, 2004:412-418.
[16] KOULOUMPIS E, WILSON T, MOORE J D. Twitter sentiment snalysis:the good the bad and the OMG![C]//International conference on weblogs and social media. Barcelona:AAAI Press, 2011:538-541.
[17] BOLLEN J, MAO H, PEPE A. Modeling public mood and emotion:Twitter sentiment and socio-economic phenomena[C]//International conference on weblogs and social media. Barcelona:AAAI Press, 2011:450-453.
[18] LIU K, XU L, ZHAO J. Opinion target extraction using word-based translation model[C]//Empirical methods in natural language processing. Stroudsburg:Association for Computational Linguistics, 2012:1346-1356.
[19] 王晰巍, 张柳, 文晴, 等. 基于贝叶斯模型的移动环境下网络舆情用户情感演化研究——以新浪微博"里约奥运会中国女排夺冠"话题为例[J]. 情报学报, 2018, 37(12):1241-1248.
[20] 廖海涵, 王曰芬, 关鹏. 微博舆情传播周期中不同传播者的主题挖掘与观点识别[J]. 图书情报工作, 2018, 62(19):77-85.
