基于情感计算的政务微博情绪话语管理研究  

An exploration of discursive emotion management in governmental microblogs based on sentiment analysis

作　　者：李捷 袁周敏[2] Li Jie;Yuan Zhoumin

机构地区：[1]南京大学大学外语部 [2]南京邮电大学外国语学院

出　　处：《外语教学》2023年第5期47-52,共6页Foreign Language Education

基　　金：国家社科基金重大项目“网络空间社会治理语言问题研究”(项目编号:20&ZD299);国家社科基金重点项目“回应突发舆情的政务微博话语及政务语言能力建设研究”(项目编号:20AYY009)的阶段性研究成果。

摘　　要：目前学界对于政务新媒体话语和情绪管理的关注日益增长,但尚未有实证研究考察危机语境下如何通过政务新媒体话语管理公众情绪。本研究利用Python情感分析筛选出突发公共事件中引发积极评论的微博,对其博文开展情绪管理分析和话语策略分析。研究结果表明,评论积极的政务微博主要在建立个人关联、展现人文关怀和凸显积极向上这三个维度上管理公众情绪,使用的话语策略包括实施特定的言语行为、使用亲属泛化称呼语、使用直接引语和融合多模态语言信息。另外,通过定量统计和分析发现,针对不同的事件类型政务微博会选择不同的情绪管理维度和话语策略。本研究在理论上促进了新媒体话语研究和情绪管理研究的跨学科融合,在实践上可以为我国政务新媒体的语言能力建设提供参考。Scholars have been increasingly interested in governmental new media discourse and emotion management,but no empirical studies have been conducted to examine how governmental new media discourse is drawn upon to manage the emotion of the public in the crisis context.The present study identifies the microblogs receiving positive comments by Python sentiment analysis,and analyzes the microblog texts in terms of emotion management and discursive strategy.The findings reveal that the microblogs receiving positive comments mainly manage the emotion of the public in such three dimensions as building personal relations,demonstrating humanistic care,and highlighting positivity,and employ such discursive strategies as using particular speech acts,employing generalized kinship address terms,adopting direct speech,and blending multimodal information.Moreover,the results of our quantitative analysis indicate that the dimensions of the emotion management and the discursive strategies in these microblogs vary according to the types of the public emergencies.Our study theoretically improves the interdisciplinary integration between the research on new media discourse and that on emotion management,and practically provides implications to the development of the language competence of China's governmental new media.

关 键 词：政务新媒体话语 政务微博 情绪管理 话语策略 情感分析 

分 类 号：H030[语言文字—语言学]