Real-time urban rainstorm and waterlogging disaster detection by Weibo users

Abstract

Abstract. With the process of urbanization in China, the urban waterlogging caused by rainstorms occurs frequently and often leads to serious damage to the natural environment, human life, and the city economy. Rapid detection of rainstorm and urban waterlogging disasters is an essential step to minimize these losses. Weibo, a popular microblog service in China, can provide many real-time Weibo posts for rapid detection. In this paper, we propose a method to identify microblogs with rainstorm and waterlogging information and apply them to waterlogging risk assessment. After pre-processing the microblog texts, we evaluate the performance of clustering (k-means) and classification (support vector machine, SVM) algorithms in the classification task. Apart from word vector features, we also introduce sentiment and publisher features for more real-time and accurate results. Furthermore, we build a waterlogging intensity dictionary to assess the waterlogging risk from the Weibo texts and produce a risk map with ArcGIS. To examine the efficacy of this approach, we collect Weibo data from two rainstorms and waterlogging disasters in Beijing city as examples. The results indicate that the SVM algorithm can be applied for a real-time rainstorm and waterlogging information detection. Compared to official-authentication and personal-certification users, the microblogs posted by general users can better indicate the intensity and timing of rainstorms. The location of waterlogging points is consistent with the risk assessment results, which proves our proposed risk assessment method can be used as a reference for timely emergency response.