观点多元化视角下的社交媒体群体极化测量与可视化研究  

Measurement and Visualization of Group Polarization on Social Media from the Perspective of Opinion Diversity

作　　者：任明[1] 陈慧铷 许光 Ren Ming

机构地区：[1]中国人民大学信息资源管理学院,北京100872

出　　处：《情报理论与实践》2023年第1期100-107,共8页Information Studies:Theory & Application

基　　金：中国人民大学重大项目“政府大数据治理与综合利用的标准化协同管理体系构建与应用研究”的成果之一,项目编号:21XNL019。

摘　　要：[目的/意义]社会热点话题常引发大众的高度关注和热议,在匿名化、高互动性、公共性的社交媒体中可能引发群体极化的现象。文章从观点多元化的视角来理解群体极化现象,根据各种观点的群体相对规模的变化趋势来量化群体极化。[方法/过程]将群体极化看作包括若干阶段的动态过程,计算各阶段的观点离散程度,并对观点离散程度序列进行回归分析。如果回归系数显著,系数为正说明存在群体极化,系数为负说明存在去极化;如果回归系数不显著,则不能判定是否出现群体极化。同时文章给出了基于文本挖掘进行社交媒体极化测量和可视化的方法。[结果/结论]以微博平台上“乃万音乐节发言”引发的男女平等的热点话题为例,使用文章提出的方法进行分析和可视化,发现在该话题的讨论中存在去极化现象。[Purpose/significance] Social hot topics often attract high attention and discussion among the public, and may lead to group polarization in social media that is characterized by anonymity, high interaction and publicity.This paper views group polarization from the perspective of diversity of opinions, and proposes to measure group polarization based on the trend of the relative sizes of various opinions.[Method/process] Group polarization is regarded as a dynamic process.The opinion dispersion degree at each stage is computed, and the sequence of opinion dispersion degree is processed by regression analysis.It can be determined that there exists group polarization if the regression coefficient is significant and positive;there exists depolarization if the regression coefficient is significant and negative;it cannot be determined there exists polarization if the regression coefficient is not significant.This paper also presents the method of measuring and visualizing group polarization on social media based on text mining.[Result/conclusion] This paper takes the discussions on gender equality caused by “Nineone’s Speech on Music Festival” on Sina Weibo as an example to analyze and visualize the group polarization in these discussions, which shows there is no group polarization.

关 键 词：群体极化 社交媒体 观点多元化 文本挖掘 可视化 

分 类 号：C912.64[经济管理] G206[社会学]