Browse by author
Lookup NU author(s): Professor Cheng Chin
This work is licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0).
The southeastern coast of China suffers many typhoon disasters every year, causing hugecasualties and economic losses. In addition, collecting statistics on typhoon disaster situations ishard work for the government. At the same time, near-real-time disaster-related information can beobtained on developed social media platforms like Twitter and Weibo. Many cases have proved thatcitizens are able to organize themselves promptly on the spot, and begin to share disasterinformation when a disaster strikes, producing massive VGI (volunteered geographic information)about the disaster situation, which could be valuable for disaster response if this VGI could beexploited efficiently and properly. However, this social media information has features such as largequantity, high noise, and unofficial modes of expression that make it difficult to obtain usefulinformation. In order to solve this problem, we first designed a new classification system based onthe characteristics of social medial data like Sina Weibo data, and made a microblogging dataset oftyphoon damage with according category labels. Secondly, we used this social medial dataset totrain the deep learning model, and constructed a typhoon disaster mining model based on a deeplearning network, which could automatically extract information about the disaster situation. Themodel is different from the general classification system in that it automatically selected microblogsrelated to disasters from a large number of microblog data, and further subdivided them intodifferent types of disasters to facilitate subsequent emergency response and loss estimation. Theadvantages of the model included a wide application range, high reliability, strong pertinence andfast speed. The research results of this thesis provide a new approach to typhoon disaster assessmentin the southeastern coastal areas of China, and provide the necessary information for theauthoritative information acquisition channel.
Author(s): Yu J, Zhao Q, Chin CS
Publication type: Article
Publication status: Published
Journal: Journal of Marine Science and Engineering
Year: 2019
Volume: 7
Issue: 138
Pages: 1-16
Print publication date: 12/09/2019
Online publication date: 13/09/2019
Acceptance date: 09/09/2019
Date deposited: 04/10/2019
ISSN (electronic): 2077-1312
Publisher: MDPI
URL: https://doi.org/10.3390/jmse7090318
DOI: 10.3390/jmse7090318
Altmetrics provided by Altmetric
