A Text Mining Framework for the Classification and Prioritization of Disaster-Related Tweets for Disaster Response

Raymond Freth Lagria Department of Industrial Engineering and Operations Research College of Engineering, University of the Philippines Diliman Quezon City, Philippines
Eugene Rex Jalao Department of Industrial Engineering and Operations Research College of Engineering, University of the Philippines Diliman Quezon City, Philippines
Joanna Resurreccion Department of Industrial Engineering and Operations Research College of Engineering, University of the Philippines Diliman Quezon City, Philippines

Abstract

Disasters have enormously disrupted the normal way of life of countries around the world and the Philippines is one of these countries. It is one of the most badly hit by disasters every year and due to its lower coping capabilities, it has constantly been ranked in the top 10 of the World Risk Index. This paper proposes a text mining framework that classifies and prioritizes disaster-related social media data, particularly Twitter tweets for the use of disaster managers for disaster response decision making. Validation of the framework during the classification stage resulted in an average of 90.67%, 99.25%, and 72.84% recall, on the test cases pertaining to training data and two different typhoon datasets. The prioritization module also prioritized tweets that were deemed urgent indicating the need for immediate response or attention.

Keywords — classification, twitter, text mining, disaster management, disaster response

Published

2023-09-04

Issue

Vol 43 No 1 (2022)

Section

Articles