No Cover Image

Journal article 547 views 128 downloads

Multi-Channel Convolutional Neural Network for the Identification of Eyewitness Tweets of Disaster

Abhinav Kumar Orcid Logo, Jyoti Prakash Singh, Nripendra P. Rana, Yogesh Dwivedi

Information Systems Frontiers, Volume: 25, Issue: 4, Pages: 1589 - 1604

Swansea University Author: Yogesh Dwivedi

Check full text

DOI (Published version): 10.1007/s10796-022-10309-x

Abstract

During a disaster, a large number of disaster-related social media posts are widely disseminated. Only a small percentage of disaster-related information is posted by eyewitnesses. The post of a disaster eyewitness offers an accurate depiction of the disaster. Therefore, the information posted by th...

Full description

Published in: Information Systems Frontiers
ISSN: 1387-3326 1572-9419
Published: Springer Science and Business Media LLC 2023
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa60323
first_indexed 2022-06-25T19:46:52Z
last_indexed 2024-11-14T12:17:07Z
id cronfa60323
recordtype SURis
fullrecord <?xml version="1.0"?><rfc1807><datestamp>2023-09-05T12:56:16.4350139</datestamp><bib-version>v2</bib-version><id>60323</id><entry>2022-06-25</entry><title>Multi-Channel Convolutional Neural Network for the Identification of Eyewitness Tweets of Disaster</title><swanseaauthors><author><sid>d154596e71b99ad1285563c8fdd373d7</sid><firstname>Yogesh</firstname><surname>Dwivedi</surname><name>Yogesh Dwivedi</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2022-06-25</date><abstract>During a disaster, a large number of disaster-related social media posts are widely disseminated. Only a small percentage of disaster-related information is posted by eyewitnesses. The post of a disaster eyewitness offers an accurate depiction of the disaster. Therefore, the information posted by the eyewitness is preferred over the other source of information as it is more effective at helping organize rescue and relief operations and potentially saving lives. In this work, we propose a multi-channel convolutional neural network (MCNN) that uses three different word-embedding vectors together to classify disaster-related tweets into eyewitness, non-eyewitness, and don&#x2019;t know classes. We compared the performance of the proposed multi-channel convolutional neural network with several attention-based deep-learning models and conventional machine learning-models such as recurrent neural network, gated recurrent unit, Long-Short-Term-Memory, convolutional neural network, logistic regression, support vector machine, and gradient boosting. The proposed multi-channel convolutional neural network achieved an F1-score of 0.84, 0.88, 0.84, and 0.86 with four disaster-related datasets of floods, earthquakes, hurricanes, and wildfires, respectively. The experimental results show that the training MCNN model with different word embedding together performs better than the conventional machine-learning models and several other deep-learning models.</abstract><type>Journal Article</type><journal>Information Systems Frontiers</journal><volume>25</volume><journalNumber>4</journalNumber><paginationStart>1589</paginationStart><paginationEnd>1604</paginationEnd><publisher>Springer Science and Business Media LLC</publisher><placeOfPublication/><isbnPrint/><isbnElectronic/><issnPrint>1387-3326</issnPrint><issnElectronic>1572-9419</issnElectronic><keywords>Disaster; Eyewitness tweets; Informative contents; Multi-channel convolutional neural network; Recurrent neural network</keywords><publishedDay>1</publishedDay><publishedMonth>8</publishedMonth><publishedYear>2023</publishedYear><publishedDate>2023-08-01</publishedDate><doi>10.1007/s10796-022-10309-x</doi><url>http://dx.doi.org/10.1007/s10796-022-10309-x</url><notes/><college>COLLEGE NANME</college><CollegeCode>COLLEGE CODE</CollegeCode><institution>Swansea University</institution><apcterm>Other</apcterm><funders/><projectreference/><lastEdited>2023-09-05T12:56:16.4350139</lastEdited><Created>2022-06-25T20:44:39.4356829</Created><path><level id="1">Faculty of Humanities and Social Sciences</level><level id="2">School of Management - Business Management</level></path><authors><author><firstname>Abhinav</firstname><surname>Kumar</surname><orcid>0000-0001-9367-7069</orcid><order>1</order></author><author><firstname>Jyoti Prakash</firstname><surname>Singh</surname><order>2</order></author><author><firstname>Nripendra P.</firstname><surname>Rana</surname><order>3</order></author><author><firstname>Yogesh</firstname><surname>Dwivedi</surname><order>4</order></author></authors><documents><document><filename>60323__24396__3b2992638f01412fb57a83516ed49165.pdf</filename><originalFilename>Eyewitness_ISF_R5_English_Proofread.pdf</originalFilename><uploaded>2022-06-25T20:46:40.2399241</uploaded><type>Output</type><contentLength>423437</contentLength><contentType>application/pdf</contentType><version>Accepted Manuscript</version><cronfaStatus>true</cronfaStatus><embargoDate>2023-07-15T00:00:00.0000000</embargoDate><copyrightCorrect>true</copyrightCorrect><language>eng</language></document></documents><OutputDurs/></rfc1807>
spelling 2023-09-05T12:56:16.4350139 v2 60323 2022-06-25 Multi-Channel Convolutional Neural Network for the Identification of Eyewitness Tweets of Disaster d154596e71b99ad1285563c8fdd373d7 Yogesh Dwivedi Yogesh Dwivedi true false 2022-06-25 During a disaster, a large number of disaster-related social media posts are widely disseminated. Only a small percentage of disaster-related information is posted by eyewitnesses. The post of a disaster eyewitness offers an accurate depiction of the disaster. Therefore, the information posted by the eyewitness is preferred over the other source of information as it is more effective at helping organize rescue and relief operations and potentially saving lives. In this work, we propose a multi-channel convolutional neural network (MCNN) that uses three different word-embedding vectors together to classify disaster-related tweets into eyewitness, non-eyewitness, and don’t know classes. We compared the performance of the proposed multi-channel convolutional neural network with several attention-based deep-learning models and conventional machine learning-models such as recurrent neural network, gated recurrent unit, Long-Short-Term-Memory, convolutional neural network, logistic regression, support vector machine, and gradient boosting. The proposed multi-channel convolutional neural network achieved an F1-score of 0.84, 0.88, 0.84, and 0.86 with four disaster-related datasets of floods, earthquakes, hurricanes, and wildfires, respectively. The experimental results show that the training MCNN model with different word embedding together performs better than the conventional machine-learning models and several other deep-learning models. Journal Article Information Systems Frontiers 25 4 1589 1604 Springer Science and Business Media LLC 1387-3326 1572-9419 Disaster; Eyewitness tweets; Informative contents; Multi-channel convolutional neural network; Recurrent neural network 1 8 2023 2023-08-01 10.1007/s10796-022-10309-x http://dx.doi.org/10.1007/s10796-022-10309-x COLLEGE NANME COLLEGE CODE Swansea University Other 2023-09-05T12:56:16.4350139 2022-06-25T20:44:39.4356829 Faculty of Humanities and Social Sciences School of Management - Business Management Abhinav Kumar 0000-0001-9367-7069 1 Jyoti Prakash Singh 2 Nripendra P. Rana 3 Yogesh Dwivedi 4 60323__24396__3b2992638f01412fb57a83516ed49165.pdf Eyewitness_ISF_R5_English_Proofread.pdf 2022-06-25T20:46:40.2399241 Output 423437 application/pdf Accepted Manuscript true 2023-07-15T00:00:00.0000000 true eng
title Multi-Channel Convolutional Neural Network for the Identification of Eyewitness Tweets of Disaster
spellingShingle Multi-Channel Convolutional Neural Network for the Identification of Eyewitness Tweets of Disaster
Yogesh Dwivedi
title_short Multi-Channel Convolutional Neural Network for the Identification of Eyewitness Tweets of Disaster
title_full Multi-Channel Convolutional Neural Network for the Identification of Eyewitness Tweets of Disaster
title_fullStr Multi-Channel Convolutional Neural Network for the Identification of Eyewitness Tweets of Disaster
title_full_unstemmed Multi-Channel Convolutional Neural Network for the Identification of Eyewitness Tweets of Disaster
title_sort Multi-Channel Convolutional Neural Network for the Identification of Eyewitness Tweets of Disaster
author_id_str_mv d154596e71b99ad1285563c8fdd373d7
author_id_fullname_str_mv d154596e71b99ad1285563c8fdd373d7_***_Yogesh Dwivedi
author Yogesh Dwivedi
author2 Abhinav Kumar
Jyoti Prakash Singh
Nripendra P. Rana
Yogesh Dwivedi
format Journal article
container_title Information Systems Frontiers
container_volume 25
container_issue 4
container_start_page 1589
publishDate 2023
institution Swansea University
issn 1387-3326
1572-9419
doi_str_mv 10.1007/s10796-022-10309-x
publisher Springer Science and Business Media LLC
college_str Faculty of Humanities and Social Sciences
hierarchytype
hierarchy_top_id facultyofhumanitiesandsocialsciences
hierarchy_top_title Faculty of Humanities and Social Sciences
hierarchy_parent_id facultyofhumanitiesandsocialsciences
hierarchy_parent_title Faculty of Humanities and Social Sciences
department_str School of Management - Business Management{{{_:::_}}}Faculty of Humanities and Social Sciences{{{_:::_}}}School of Management - Business Management
url http://dx.doi.org/10.1007/s10796-022-10309-x
document_store_str 1
active_str 0
description During a disaster, a large number of disaster-related social media posts are widely disseminated. Only a small percentage of disaster-related information is posted by eyewitnesses. The post of a disaster eyewitness offers an accurate depiction of the disaster. Therefore, the information posted by the eyewitness is preferred over the other source of information as it is more effective at helping organize rescue and relief operations and potentially saving lives. In this work, we propose a multi-channel convolutional neural network (MCNN) that uses three different word-embedding vectors together to classify disaster-related tweets into eyewitness, non-eyewitness, and don’t know classes. We compared the performance of the proposed multi-channel convolutional neural network with several attention-based deep-learning models and conventional machine learning-models such as recurrent neural network, gated recurrent unit, Long-Short-Term-Memory, convolutional neural network, logistic regression, support vector machine, and gradient boosting. The proposed multi-channel convolutional neural network achieved an F1-score of 0.84, 0.88, 0.84, and 0.86 with four disaster-related datasets of floods, earthquakes, hurricanes, and wildfires, respectively. The experimental results show that the training MCNN model with different word embedding together performs better than the conventional machine-learning models and several other deep-learning models.
published_date 2023-08-01T02:29:16Z
_version_ 1821370804095418368
score 11.04748

Similar Items