基于语料库的新冠疫情网络谣言分析

A Corpus-based Analysis of COVID-19 Internet Rumors

对基于具体事件的网络谣言进行语言特征分析既有助于深入发现网络谣言的语言特征,也有利于对今后类似事件中网络谣言的识别和预防。在自建新冠疫情网络谣言语料库的基础上,文章通过提取主题词对比分析的方法总结了新冠谣言的如下特征:主题词语义集中化、名词使用抽象化、动词施动者以民众为主、祈使词语具有强制意味。通过与人民日报官微语料的对比,发现官微语料语义更加明确、细节性更强;与官微相比,谣言具有以下特征:更多使用祈使类词语、对物资的关注更多、在时间设定上具有较强随意性、好用强制意味的词语来增强内容的权威感。总体上看,新冠网络谣言的特点集中表现在“多使用民众视角、语义具有强制性”两方面。

Analyzing the linguistic features of network rumors based on specific events is not only helpful to find out the linguistic features of network rumors,but also helpful to identify and prevent network rumors in similar events in the future.During the outbreak of CO VID-19,a lot of Internet rumors emerged.Through the comparative analysis of the subject words extracted from theself-established rumor corpus,the paper finds that the related rumors have the characteristics of the semantic centralization of the subject words,abstract use of the nouns,making the masses as the main verb actors,and the mandatory meaning of the imperative words.By contrast with the corpus of People's Daily official platforms,it lacks clear semantics and strong details.Moreover,the rumors tend to use more imperative words and focuses more on materials,showing quite strong randomness on the time seting and using the words with compulsory meaning to strengthen the authority of the content.Hence,these confirm its features of being based on the masses'perspective and mandatory semantics.