Identifying informative messages in disaster events using Convolutional Neural Networks

Cornelia Caragea, Adrian Silvescu, Andrea H. Tapia

Research output: Chapter in Book/Report/Conference proceedingConference contribution

24 Citations (Scopus)

Abstract

Social media is a vital source of information during any major event, especially natural disasters. Data produced through social networking sites is seen as ubiquitous, rapid and accessible, and it is believed to empower average citizens to become more situationally aware during disasters and coordinate to help themselves. However, with the exponential increase in the volume of social media data, so comes the increase in data that are irrelevant to a disaster, thus, diminishing peoples ability to find the information that they need in order to organize relief efforts, find help, and potentially save lives. In this paper, we present an approach to identifying informative messages in social media streams during disaster events. Our approach is based on Convolutional Neural Networks and shows significant improvement in performance over models that use the "bag of words" and n-grams as features on several datasets of messages from flooding events.

Original languageEnglish (US)
Title of host publicationISCRAM 2016 Conference Proceedings - 13th International Conference on Information Systems for Crisis Response and Management
EditorsPedro Antunes, Victor Amadeo Banuls Silvera, Joao Porto de Albuquerque, Kathleen Ann Moore, Andrea H. Tapia
PublisherInformation Systems for Crisis Response and Management, ISCRAM
ISBN (Electronic)9788460879848
StatePublished - Jan 1 2016
Event13th International Conference on Information Systems for Crisis Response and Management, ISCRAM 2016 - Rio de Janeiro, Brazil
Duration: May 22 2016May 25 2016

Publication series

NameProceedings of the International ISCRAM Conference
ISSN (Electronic)2411-3387

Other

Other13th International Conference on Information Systems for Crisis Response and Management, ISCRAM 2016
CountryBrazil
CityRio de Janeiro
Period5/22/165/25/16

Fingerprint

Disasters
Neural networks
Disaster
Social media
Social networking sites
Sources of information
Natural disasters
Flooding

All Science Journal Classification (ASJC) codes

  • Computer Networks and Communications
  • Information Systems
  • Information Systems and Management
  • Electrical and Electronic Engineering

Cite this

Caragea, C., Silvescu, A., & Tapia, A. H. (2016). Identifying informative messages in disaster events using Convolutional Neural Networks. In P. Antunes, V. A. Banuls Silvera, J. Porto de Albuquerque, K. A. Moore, & A. H. Tapia (Eds.), ISCRAM 2016 Conference Proceedings - 13th International Conference on Information Systems for Crisis Response and Management (Proceedings of the International ISCRAM Conference). Information Systems for Crisis Response and Management, ISCRAM.
Caragea, Cornelia ; Silvescu, Adrian ; Tapia, Andrea H. / Identifying informative messages in disaster events using Convolutional Neural Networks. ISCRAM 2016 Conference Proceedings - 13th International Conference on Information Systems for Crisis Response and Management. editor / Pedro Antunes ; Victor Amadeo Banuls Silvera ; Joao Porto de Albuquerque ; Kathleen Ann Moore ; Andrea H. Tapia. Information Systems for Crisis Response and Management, ISCRAM, 2016. (Proceedings of the International ISCRAM Conference).
@inproceedings{1245d1a164ef4b94aaabc448d40aac22,
title = "Identifying informative messages in disaster events using Convolutional Neural Networks",
abstract = "Social media is a vital source of information during any major event, especially natural disasters. Data produced through social networking sites is seen as ubiquitous, rapid and accessible, and it is believed to empower average citizens to become more situationally aware during disasters and coordinate to help themselves. However, with the exponential increase in the volume of social media data, so comes the increase in data that are irrelevant to a disaster, thus, diminishing peoples ability to find the information that they need in order to organize relief efforts, find help, and potentially save lives. In this paper, we present an approach to identifying informative messages in social media streams during disaster events. Our approach is based on Convolutional Neural Networks and shows significant improvement in performance over models that use the {"}bag of words{"} and n-grams as features on several datasets of messages from flooding events.",
author = "Cornelia Caragea and Adrian Silvescu and Tapia, {Andrea H.}",
year = "2016",
month = "1",
day = "1",
language = "English (US)",
series = "Proceedings of the International ISCRAM Conference",
publisher = "Information Systems for Crisis Response and Management, ISCRAM",
editor = "Pedro Antunes and {Banuls Silvera}, {Victor Amadeo} and {Porto de Albuquerque}, Joao and Moore, {Kathleen Ann} and Tapia, {Andrea H.}",
booktitle = "ISCRAM 2016 Conference Proceedings - 13th International Conference on Information Systems for Crisis Response and Management",

}

Caragea, C, Silvescu, A & Tapia, AH 2016, Identifying informative messages in disaster events using Convolutional Neural Networks. in P Antunes, VA Banuls Silvera, J Porto de Albuquerque, KA Moore & AH Tapia (eds), ISCRAM 2016 Conference Proceedings - 13th International Conference on Information Systems for Crisis Response and Management. Proceedings of the International ISCRAM Conference, Information Systems for Crisis Response and Management, ISCRAM, 13th International Conference on Information Systems for Crisis Response and Management, ISCRAM 2016, Rio de Janeiro, Brazil, 5/22/16.

Identifying informative messages in disaster events using Convolutional Neural Networks. / Caragea, Cornelia; Silvescu, Adrian; Tapia, Andrea H.

ISCRAM 2016 Conference Proceedings - 13th International Conference on Information Systems for Crisis Response and Management. ed. / Pedro Antunes; Victor Amadeo Banuls Silvera; Joao Porto de Albuquerque; Kathleen Ann Moore; Andrea H. Tapia. Information Systems for Crisis Response and Management, ISCRAM, 2016. (Proceedings of the International ISCRAM Conference).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Identifying informative messages in disaster events using Convolutional Neural Networks

AU - Caragea, Cornelia

AU - Silvescu, Adrian

AU - Tapia, Andrea H.

PY - 2016/1/1

Y1 - 2016/1/1

N2 - Social media is a vital source of information during any major event, especially natural disasters. Data produced through social networking sites is seen as ubiquitous, rapid and accessible, and it is believed to empower average citizens to become more situationally aware during disasters and coordinate to help themselves. However, with the exponential increase in the volume of social media data, so comes the increase in data that are irrelevant to a disaster, thus, diminishing peoples ability to find the information that they need in order to organize relief efforts, find help, and potentially save lives. In this paper, we present an approach to identifying informative messages in social media streams during disaster events. Our approach is based on Convolutional Neural Networks and shows significant improvement in performance over models that use the "bag of words" and n-grams as features on several datasets of messages from flooding events.

AB - Social media is a vital source of information during any major event, especially natural disasters. Data produced through social networking sites is seen as ubiquitous, rapid and accessible, and it is believed to empower average citizens to become more situationally aware during disasters and coordinate to help themselves. However, with the exponential increase in the volume of social media data, so comes the increase in data that are irrelevant to a disaster, thus, diminishing peoples ability to find the information that they need in order to organize relief efforts, find help, and potentially save lives. In this paper, we present an approach to identifying informative messages in social media streams during disaster events. Our approach is based on Convolutional Neural Networks and shows significant improvement in performance over models that use the "bag of words" and n-grams as features on several datasets of messages from flooding events.

UR - http://www.scopus.com/inward/record.url?scp=85015806207&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85015806207&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:85015806207

T3 - Proceedings of the International ISCRAM Conference

BT - ISCRAM 2016 Conference Proceedings - 13th International Conference on Information Systems for Crisis Response and Management

A2 - Antunes, Pedro

A2 - Banuls Silvera, Victor Amadeo

A2 - Porto de Albuquerque, Joao

A2 - Moore, Kathleen Ann

A2 - Tapia, Andrea H.

PB - Information Systems for Crisis Response and Management, ISCRAM

ER -

Caragea C, Silvescu A, Tapia AH. Identifying informative messages in disaster events using Convolutional Neural Networks. In Antunes P, Banuls Silvera VA, Porto de Albuquerque J, Moore KA, Tapia AH, editors, ISCRAM 2016 Conference Proceedings - 13th International Conference on Information Systems for Crisis Response and Management. Information Systems for Crisis Response and Management, ISCRAM. 2016. (Proceedings of the International ISCRAM Conference).