Metro plays an important role in middle and long-distance travel in large cities. However, aging equipment and maintenance problems may lead to metro accidents. In this paper, social media data are crawled from both the official and personal microblog to study the severity of accidents and subjective attitude of passengers. The relationship between metro ridership and accident is investigated. In addition, natural language processing (NLP) technology is applied to analyze the microblog text, so that public attitudes and responses to metro accidents can be identified. Through this method, the direct impact of metro accidents can be measured by easily-obtained and time-sensitive social media data. Moreover, this method is also helpful to evaluate the effect of emergency measures for the metro accident.
展开▼
