海外自媒体中涉华舆情传播机制的大数据分析——基于Reddit平台的海量舆情信息

发布时间：2018-04-28 16:31

本文选题：谷歌云计算 + 大数据　；参考：《学术论坛》2017年03期

【摘要】：文章基于谷歌云计算平台,以国际自媒体平台Reddit 2007—2015年间涉华信息所构成的舆情大数据为分析对象,透视国际自媒体领域涉华舆情的传播机制,并试图提出相应的理论解释。通过卷积神经网络等文本挖掘技术法,文章对涉华舆情的两个方面,即63万条涉华新闻帖子以及540万条涉华评论的情感倾向进行分类。研究发现,Reddit平台上涉华信息的传播并不是随机分布的,消极负面帖子明显多于积极正面帖子,相对于客观中立和积极的帖子,消极负面帖子具有更高的被关注度和更强的传播力;在受众所发出的涉华评论中,其情感倾向的分布也呈现出负面主导的突出特征,即客观中立和积极正面评论被淹没在消极负面评论的汪洋大海中。涉华舆情的这样一种传播模式与国际自媒体的
[Abstract]:Based on the cloud computing platform of Google, this paper analyzes the public opinion big data made up of the information about China from 2007 to 2015 on the international self-media platform Reddit, and attempts to put forward the corresponding theoretical explanation on the transmission mechanism of the public opinion concerning China in the field of international self-media. By means of convolution neural network and other text mining techniques, this paper classifies two aspects of public opinion concerning China, that is, 630000 Chinese news posts and 5.4 million comments on China. The study found that the dissemination of Chinese information on Reddit platform is not random distribution, negative posts are obviously more than positive posts, compared with objective neutral and positive posts, Negative and negative posts have a higher degree of attention and a stronger power of communication, and the distribution of their emotional tendencies also shows a negative dominant feature in the comments issued by the audience concerning China. Objective neutral and positive comments are submerged in the sea of negative comments. Such a mode of dissemination of public opinion concerning China and the international media
【作者单位】：武汉大学社会学系;武汉大学国际软件学院;
【基金】：国家社科基金重大项目“大数据时代计算社会科学的产生、现状与发展趋势研究”(16ZDA086) 中国博士后科学基金第60批面上资助项目(2016M602372) 第10批特别资助项目(2017T100587) 武汉大学自主科研项目“‘一带一路’沿线文化认同与文化隔阂的大数据分析”(2017QN047)等项目资助
【分类号】：G206

【相似文献】