Sentiment Analysis on Evacuation and Relief Operation in the Philippines

Abstract

Individual sentiments on current certain issues were expressed through social media as well as to get the pulse of the majority population in certain circumstances, specifically on disaster management issues. To conduct a sentiment analysis of the disaster evacuation and relief operations in the Philippines, the researchers used Twitter social media. The qualitative research methods and sentiment analysis were used to collect and analyze the data. The GetOldTweets application was used to collect the tweets related to disaster relief and evacuation from the year 2013 to the year 2017 and serves as the data sets for the analysis of sentiments. Social media posts written in English and Filipino tweets were analyzed by using the Waikato Environment for Knowledge Analysis (WEKA) tools to manually annotate positive or negative tweets and validate classification model output using stratified cross validation and support vector machines (SVMs) algorithms. The manual classification model received a high percentage of correctly categorized instances using the 10-fold cross validation with the classifier SVMs. Tweets with positive and negative sentiments will significantly improve to have an effective and efficient activity for relief and evacuation. Government needs several measures to strengthen the evacuation process and the management system for relief operations. © 2020, World Academy of Research in Science and Engineering. All rights reserved.