Extraction and Visual Analysis of Negative Traffic Events from Weibo Data

Social media has an increasingly significant importance in people’s daily life during past few years. Social media data has been widely studied in a variety of disciplines for different applications, e.g., crisis management and urban planning. In this work, we focus on the extraction and analysis of traffic related events, especially negative traffic events(NTE), e.g. congestion, car accidents, from social media data. Firstly, we identify the terms related to NTEs. Secondly, based on those terms, an iterative lexiconbased text mining technique is applied to extract NTEs from social media data. Thirdly, we calculate the statistics and visualize the spatiotemporal patterns of the NTEs. A web-based interactive visualization system is developed for visual analysis of NTEs. We use one year Sina Weibo data in Shanghai as our test data and present our preliminary results.