Welcome to the InfoSci Platform
Reference Hub9
Enabling Rapid Classification of Social Media Communications During Crises

Enabling Rapid Classification of Social Media Communications During Crises

Muhammad Imran, Prasenjit Mitra, Jaideep Srivastava
Copyright: © 2016 |Volume: 8 |Issue: 3 |Pages: 17
ISSN: 1937-9390|EISSN: 1937-9420|EISBN13: 9781466690509|DOI: 10.4018/IJISCRAM.2016070101
Cite Article Cite Article

MLA

Imran, Muhammad, et al. "Enabling Rapid Classification of Social Media Communications During Crises." IJISCRAM vol.8, no.3 2016: pp.1-17. http://doi.org/10.4018/IJISCRAM.2016070101

APA

Imran, M., Mitra, P., & Srivastava, J. (2016). Enabling Rapid Classification of Social Media Communications During Crises. International Journal of Information Systems for Crisis Response and Management (IJISCRAM), 8(3), 1-17. http://doi.org/10.4018/IJISCRAM.2016070101

Chicago

Imran, Muhammad, Prasenjit Mitra, and Jaideep Srivastava. "Enabling Rapid Classification of Social Media Communications During Crises," International Journal of Information Systems for Crisis Response and Management (IJISCRAM) 8, no.3: 1-17. http://doi.org/10.4018/IJISCRAM.2016070101

Export Reference

Mendeley
Favorite Full-Issue Download

Abstract

The use of social media platforms such as Twitter by affected people during crises is considered a vital source of information for crisis response. However, rapid crisis response requires real-time analysis of online information. When a disaster happens, among other data processing techniques, supervised machine learning can help classify online information in real-time. However, scarcity of labeled data causes poor performance in machine training. Often labeled data from past event is available. Can past labeled data be reused to train classifiers? We study the usefulness of labeled data of past events. We observe the performance of our classifiers trained using different combinations of training sets obtained from past disasters. Moreover, we propose two approaches (target labeling and active learning) to boost classification performance of a learning scheme. We perform extensive experimentation on real crisis datasets and show the utility of past-labeled data to train machine learning classifiers to process sudden-onset crisis-related data in real-time.

Request Access

You do not own this content. Please login to recommend this title to your institution's librarian or purchase it from the IGI Global bookstore.