摘要
[目的/意义]旨在分析网络舆情特征,帮助企业与监管部门提高网络舆情管控与治理水平。[方法/过程]引入SnowNLP模型,以“钟薛高雪糕事件”的网络舆情为研究对象,对微博文本进行情感分析从而划分舆情演化周期,并使用Biterm Topic Model(BTM)模型进行主题提取。在此基础上构建主题-情感融合分析模型,从个人类意见领袖、信息发布者、媒体传播者三个维度分阶段对舆情热点主题和情感演化展开交叉研究,探寻不同群体视角下舆情演化的传播机制。[结果/结论]网络舆情的演化周期分为四个阶段:形成期、扩散期、爆发期、终结期,整个舆情周期较短,呈现出动态的演化阶段特点。个人类意见领袖发言自由随意,更容易煽动群众情绪;信息发布者言论影响力大,但控制舆情的预期效果不佳;媒体传播者言论缺乏应有的客观性和中立性,加速舆情发酵。从整体上看网络情绪呈现负面极化状态。
[Purpose/significance]It aims to analyze the characteristics of online public opinion and help enterprises and regulatory authorities improve the level of online public opinion control and governance.[Method/process]This paper introduces the SnowNLP model takes the online public opinion of“Zhong Xue Gao Ice Cream Event”as the research object to make sentiment analysis of the microblog text divides the evolution cycle of public opinion and uses the Biterm Topic Model(BTM)for topic extraction.On this basis this paper constructs a topic-emotion fusion analysis model and conducts a cross-sectional study on the evolution of hot topics and emotions in three dimensions:individual opinion leaders information publishers and media disseminators so as to explore the communication mechanism of public opinion evolution from different groups’perspectives.[Result/conclusion]The evolution cycle of online public opinion is divided into four stages:formation period proliferation period outbreak period and final period and the whole public opinion cycle is short showing the characteristics of dynamic evolution stages.Individual opinion leaders speak freely and casually which makes it easier to incite public sentiment;information publishers’remarks are influential but the expected effect of controlling public opinion is not ideal;media communicators’remarks lack proper objectivity and neutrality which accelerates the fermentation of public opinion.As a whole online sentiment presents a negative polarization state.
作者
杨沁蓉
朱光
雍华贵
Yang Qinrong;Zhu Guang;Yong Huagui(School of Management Science and Engineering Nanjing University of Information Science and Technology,Nanjing Jiangsu 210044;Institute of Risk Management and Emergency Decision Nanjing University of Information Science and Technology,Nanjing Jiangsu 210044)
出处
《情报探索》
2023年第8期111-119,共9页
Information Research
关键词
网络舆情
舆情演化
用户情感
主题模型
online public opinion
public opinion evolution
user sentiment
topic model
