Disaster Severity Analysis from Micro-Blog Texts Using Deep-NN

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

“…Barker et al (2019) developed a prototype of a national-scale Twitter data mining pipeline for improved stakeholder situational awareness during flooding events across Great Britain by retrieving relevant social geodata, grounded in environment data sources (flood warnings and river levels) (Barker et al, 2019). Wang et al (2012) analyzed the subject words and user sentiments of the earthquake events based on the week-long discussion on Sina Weibo after the earthquake in Japan (Wang et al, 2012). Zhang and Wang (2015) used the Shanghai Bund stampede incident as an example, according to the response time, response speed, microblog contents, and microblog interaction of the government microblog after the emergency, to analyze and evaluate the information release and response ability of the government in an emergency.…”

“…Text classification is a hot topic in the field of natural language processing (NLP) (Hu et al, 2018;Kim, 2014) and has been widely used to identify important information of interest from social media. For example, Wadawadagi and Pagi (2020) investigated the severity of disaster events from microblog messages during natural calamities and emergencies using convolutional neural networks (CNNs) and recurrent neural networks (RNNs). Their work employed a joint model to combine the features of CNNs with RNNs, taking account of the coarse-grained local features generated via CNNs and long-range dependencies learned through RNNs for analysis of small text messages (Wadawadagi and Pagi, 2020).…”

“…For example, Wadawadagi and Pagi (2020) investigated the severity of disaster events from microblog messages during natural calamities and emergencies using convolutional neural networks (CNNs) and recurrent neural networks (RNNs). Their work employed a joint model to combine the features of CNNs with RNNs, taking account of the coarse-grained local features generated via CNNs and long-range dependencies learned through RNNs for analysis of small text messages (Wadawadagi and Pagi, 2020). Also, Singh et al (2020) investigated the problem of localization using the social sensing model (Twitter) to provide an efficient, reliable, and accurate flood text classification model with minimal labeled data.…”