Online urban-waterlogging monitoring based on a recurrent neural network for classification of microblogging text

Abstract. With the global climate change and rapid urbanization, urban flood disasters spread become increasingly serious in China. Urban rainstorms waterlogging have an urgent challenge that needs to be monitored real time further predicted for improvement of urbanization construction. We trained a recurrent neural network (RNN) model classify microblogging posts related establish online monitoring system caused by disasters. manually curated more than 4400 train RNN so it can precisely identify waterlogging-related Sina Weibo timely determine waterlogging. The has been thoroughly evaluated, our experimental results showed achieved higher accuracy traditional machine learning methods, such as support vector (SVM) gradient boosting decision tree (GBDT). Furthermore, we build nationwide map based on recent 2-year data.