Enhancing Warning, Situational Awareness, Assessment and Education in Managing Emergency: Case Study of COVID-19

Abstract

The volume of network and Internet traffic is increasing extraordinarily fast daily, creating huge data. With this volume, variety, speed, and precision of data, it is hard to collect crisis information in such a massive data environment. This paper proposes a hybrid of deep convolutional neural network (CNN)-long short-term memory (LSTM)-based model to efficiently retrieve crisis information. Deep CNN is used to extract significant characteristics from multiple sources. LSTM is used to maintain long-term dependencies in extracted characteristics while preventing overfitting on recurring connections. This method has been compared to previous approaches to the performance of a publicly available dataset to demonstrate its highly satisfactory performance. This new approach allows integrating artificial intelligence technologies, deep learning and social media in managing crisis model. It is based on an extension of our previous approach namely long short-term memory-based disaster management and education: this experience forms a background for this model. It combines representation training with situational awareness and education, while retrieving template information by combining various search results from multiple sources. We have extended it to improve our managing disaster model and evaluate it in the case of the coronavirus disease 2019 (COVID-19) while achieving promising results.