期刊文献+

情感视域下社交媒体平台舆论分层与社群挖掘研究 被引量:1

Research on Social Media Platform Opinion Stratification and Community Mining Using Sentiment Measurement
原文传递
导出
摘要 网络舆论热点事件中,用户的情感信息和意见是影响事件发展方向的重要因素。社会的情感往往是社会结构和文化的产物。特定公众群体情感种类及强度的差异可以反映宏观的社会生产,进而反映社会的阶层分布。研究舆论分化现象及其情感特征,使用情感分类及强度去识别网络社群具有重要的学术意义,但目前该领域的实证研究仍待开拓。本研究以"假疫苗"事件为个案,使用情感词典评价社交媒体用户的情感类型及强度,并进行聚类分析;基于聚类的结果,结合社交媒体用户账号标签、讨论文本、行为数据,对聚类结果进行社群画像,评价聚类的效果。研究发现,基于情感的社群识别结果具备阶层差异的典型特征,群际差异明显,群内成员聚合性高。不同社群在公共事件中会呈现情感表露的差异,这种情感差异在经过社交媒体平台信息及情感的流动后,会在群体内部形成情感趋同。更细粒度的情感分类考察,能让社群的边界更加清晰,提升聚类结果的可解释性。 The users’sentiment and opinions are the important factors which influence the development of public opinion focusing on hot event in cyberspace.Social sentiments are generally produced by the social structure and culture.The sentimental difference of the types and intensity existing in different public groups can reflect macroscopic social production,which in turn reflects the social class distribution.It is of great academic significance to study the phenomenon of public opinion polarization and its sentimental characteristics,and to use sentimental classification and intensity to identify network community.But at present,the empirical research in the field isn’t well developed.Taking the incident of"fake vaccine"as a case,the sentiment dictionary was used to evaluate type and intensity of the sentiment pervading among social media users,and also cluster them.Use tags,discussion records,and behavioral data of social media user to portrait clustered community,and then evaluate clustering effects.The result of sentiment-based community cluster has the typical characteristics of class difference,obvious intergroup difference and high aggregation of within-group.Different communities behave differences,which form the sentimental convergence within the group because of the flowing of information and sentiment on social media platform,in sentimental disclosure when involving public events.The observation of finer-grained emotion classification makes the boundary of different community clearer and improve the interpretability of clustering results.
作者 刘昊 Liu Hao
出处 《中国网络传播研究》 CSSCI 2018年第2期-,共16页 China Computer-Mediated Communication Studies
基金 国家社科一般项目“基于社交媒体‘一带一路’国际传播效果评估与提升路径研究”(项目编号:18BXW027)的研究成果
关键词 情感聚类 情感词典 社会阶层 社群画像 “假疫苗” sentiment clustering sentiment dictionary social class community portrait "fake vaccine"
作者简介 刘昊,博士,四川外国语大学新闻传播学院教授,硕士生导师,研究方向为社交媒体舆情、计算传播
  • 相关文献

参考文献10

二级参考文献102

共引文献959

引证文献1

二级引证文献1

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部