Robust classification of crisis-related data on social networks using convolutional neural networks

Dat Tien Nguyen, Kamela Ali Al Mannai, Shafiq Rayhan Joty, Hassan Sajjad, Muhammad Imran, Prasenjit Mitra

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

19 Citations (Scopus)

Abstract

The role of social media, in particular microblogging platforms such as Twitter, as a conduit for actionable and tactical information during disasters is increasingly acknowledged. However, time-critical analysis of big crisis data on social media streams brings challenges to machine learning techniques, especially the ones that use supervised learning. The scarcity of labeled data, particularly in the early hours of a crisis, delays the learning process. Existing classification methods require a significant amount of labeled data specific to a particular event for training plus a lot of feature engineering to achieve best results. In this work, we introduce neural network based classification methods for identifying useful tweets during a crisis situation. At the onset of a disaster when no labeled data is available, our proposed method makes the best use of the out-of-event data and achieves good results.

Original languageEnglish
Title of host publicationProceedings of the 11th International Conference on Web and Social Media, ICWSM 2017
PublisherAAAI press
Pages632-635
Number of pages4
ISBN (Electronic)9781577357889
Publication statusPublished - 1 Jan 2017
Event11th International Conference on Web and Social Media, ICWSM 2017 - Montreal, Canada
Duration: 15 May 201718 May 2017

Other

Other11th International Conference on Web and Social Media, ICWSM 2017
CountryCanada
CityMontreal
Period15/5/1718/5/17

Fingerprint

Disasters
Neural networks
Supervised learning
Learning systems
Big data

ASJC Scopus subject areas

  • Computer Networks and Communications

Cite this

Nguyen, D. T., Al Mannai, K. A., Rayhan Joty, S., Sajjad, H., Imran, M., & Mitra, P. (2017). Robust classification of crisis-related data on social networks using convolutional neural networks. In Proceedings of the 11th International Conference on Web and Social Media, ICWSM 2017 (pp. 632-635). AAAI press.

Robust classification of crisis-related data on social networks using convolutional neural networks. / Nguyen, Dat Tien; Al Mannai, Kamela Ali; Rayhan Joty, Shafiq; Sajjad, Hassan; Imran, Muhammad; Mitra, Prasenjit.

Proceedings of the 11th International Conference on Web and Social Media, ICWSM 2017. AAAI press, 2017. p. 632-635.

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

Nguyen, DT, Al Mannai, KA, Rayhan Joty, S, Sajjad, H, Imran, M & Mitra, P 2017, Robust classification of crisis-related data on social networks using convolutional neural networks. in Proceedings of the 11th International Conference on Web and Social Media, ICWSM 2017. AAAI press, pp. 632-635, 11th International Conference on Web and Social Media, ICWSM 2017, Montreal, Canada, 15/5/17.
Nguyen DT, Al Mannai KA, Rayhan Joty S, Sajjad H, Imran M, Mitra P. Robust classification of crisis-related data on social networks using convolutional neural networks. In Proceedings of the 11th International Conference on Web and Social Media, ICWSM 2017. AAAI press. 2017. p. 632-635
Nguyen, Dat Tien ; Al Mannai, Kamela Ali ; Rayhan Joty, Shafiq ; Sajjad, Hassan ; Imran, Muhammad ; Mitra, Prasenjit. / Robust classification of crisis-related data on social networks using convolutional neural networks. Proceedings of the 11th International Conference on Web and Social Media, ICWSM 2017. AAAI press, 2017. pp. 632-635
@inproceedings{71b5a4e9da784312ac89dbf36080df28,
title = "Robust classification of crisis-related data on social networks using convolutional neural networks",
abstract = "The role of social media, in particular microblogging platforms such as Twitter, as a conduit for actionable and tactical information during disasters is increasingly acknowledged. However, time-critical analysis of big crisis data on social media streams brings challenges to machine learning techniques, especially the ones that use supervised learning. The scarcity of labeled data, particularly in the early hours of a crisis, delays the learning process. Existing classification methods require a significant amount of labeled data specific to a particular event for training plus a lot of feature engineering to achieve best results. In this work, we introduce neural network based classification methods for identifying useful tweets during a crisis situation. At the onset of a disaster when no labeled data is available, our proposed method makes the best use of the out-of-event data and achieves good results.",
author = "Nguyen, {Dat Tien} and {Al Mannai}, {Kamela Ali} and {Rayhan Joty}, Shafiq and Hassan Sajjad and Muhammad Imran and Prasenjit Mitra",
year = "2017",
month = "1",
day = "1",
language = "English",
pages = "632--635",
booktitle = "Proceedings of the 11th International Conference on Web and Social Media, ICWSM 2017",
publisher = "AAAI press",

}

TY - GEN

T1 - Robust classification of crisis-related data on social networks using convolutional neural networks

AU - Nguyen, Dat Tien

AU - Al Mannai, Kamela Ali

AU - Rayhan Joty, Shafiq

AU - Sajjad, Hassan

AU - Imran, Muhammad

AU - Mitra, Prasenjit

PY - 2017/1/1

Y1 - 2017/1/1

N2 - The role of social media, in particular microblogging platforms such as Twitter, as a conduit for actionable and tactical information during disasters is increasingly acknowledged. However, time-critical analysis of big crisis data on social media streams brings challenges to machine learning techniques, especially the ones that use supervised learning. The scarcity of labeled data, particularly in the early hours of a crisis, delays the learning process. Existing classification methods require a significant amount of labeled data specific to a particular event for training plus a lot of feature engineering to achieve best results. In this work, we introduce neural network based classification methods for identifying useful tweets during a crisis situation. At the onset of a disaster when no labeled data is available, our proposed method makes the best use of the out-of-event data and achieves good results.

AB - The role of social media, in particular microblogging platforms such as Twitter, as a conduit for actionable and tactical information during disasters is increasingly acknowledged. However, time-critical analysis of big crisis data on social media streams brings challenges to machine learning techniques, especially the ones that use supervised learning. The scarcity of labeled data, particularly in the early hours of a crisis, delays the learning process. Existing classification methods require a significant amount of labeled data specific to a particular event for training plus a lot of feature engineering to achieve best results. In this work, we introduce neural network based classification methods for identifying useful tweets during a crisis situation. At the onset of a disaster when no labeled data is available, our proposed method makes the best use of the out-of-event data and achieves good results.

UR - http://www.scopus.com/inward/record.url?scp=85029455949&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85029455949&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:85029455949

SP - 632

EP - 635

BT - Proceedings of the 11th International Conference on Web and Social Media, ICWSM 2017

PB - AAAI press

ER -