摘要
[研究目的]对美国大选社交机器人涉华舆论道德基础进行深入剖析,把握社交机器人涉华舆论操控,加强国际社交媒体时代国际涉华舆论引导。[研究方法]使用Python爬取涉华议题推文。利用社交机器人识别指数工具Botometer识别社交机器人账号,建立社交机器人原创推文语料库。借助道德基础词典,通过谷歌新闻词向量模型获取美国大选涉华议题的道德基础词向量。[研究结论]“忠诚/背叛”“权威/颠覆”和“关爱/伤害”成为美国大选涉华计算宣传的主要道德基础,强调“忠诚”“权威”和“伤害”。涉华计算宣传制造极化舆论,并呈现明显的共和党态度偏向,煽动拜登忠于中国,是“中国拜登”(Chinajoe),如果拜登当选将“服从”中国,美国将面临中国“威胁”,从而“颠覆”美国“权威”,“伤害”美国利益。
[Research purpose] This paper attempts to analyze the moral foundations of China-related public opinions by social robots in the U.S.election,grasp the manipulation of China-related public opinion by social robots,and strengthen the guidance of international China-related public opinions in the international social media era.[Research method] This paper selects China-related tweets using Python,uses Botometer,a social bot identification index tool,to identify social bot accounts and build a corpus of original social bot tweets,and obtains the word vectors of China-related issues tweets in U.S.election by Google news Word2 Vec model with the help of moral foundations dictionary.[Research conclusion] "Loyalty/betrayal" "authority/subversion" and "care/harm" are the main moral foundations of China-related technical propaganda in the U.S.election,emphasizing "Loyalty","authority",and "harm".The China-related propaganda creates polarized public opinion,and presents a clear bias toward Republican attitudes,inciting Biden to be loyal to China,to be "Chinajoe",and to "obey" China if Biden is elected,and the United States will face the Chinese "threat","subverting" U.S."authority" and "harm" U.S.interests.
作者
相德宝
韦向洁
Xiang Debao;Wei Xiangjie(School of International Journalism and Communication,Beijing Foreign Studies University,Beijing 100089;School of Journalism and Communication,Shanghai International Studies University,Shanghai 201600)
出处
《情报杂志》
CSSCI
北大核心
2022年第3期108-113,共6页
Journal of Intelligence
基金
上海市社科规划年度课题“全球中国通国际社交媒体中国观、影响路径与引导策略研究”(编号:2021BXW011)
国家社会科学重大项目“中国特色对外话语体系在英语世界的译介与传播研究(1949-2019)”(编号:19ZDA339)子课题五研究成果之一。
关键词
美国大选
涉华舆情
社交机器人
计算宣传
道德基础理论
词向量
the U.S.election
China-related public opinion
social bot
calculation propaganda
moral foundation theory
word embedding
