A Study of Emergency Web Rumors in Weibo Based on LDA

Weibo is an extremely high volume platform for user to post any topic, even though the content may not be true. The abundant content created at real-time contains massive information to identify topics. Such topics can be utilized in numerous ways, such as public opinion mining, marketing and emergency information diffusion. However, in microblogging systems like Weibo, emergency and web rumors often appear together. Facing web rumors, it is difficult for the masses to distinguish between the true and the false, whose opinions are often following the crowd, causing the public opinion crisis. Models, such as LDA topic model, propose algorithms for relating collections of documents to sets of keywords that represent underly topics. Based on this approach, this paper takes the 2017 Taifu Middle School in Sichuan as an example, collects comments of the netizens in Weibo, and analyzes the topics in this emergency. This paper describes the proposed approach, do a case study and propose solutions which contribute to managing web rumors and controlling the spread of rumors in emergencies.