Natural Disaster Classification in Thai Tweets

摘要

Websites and social media are become an essential sources of information during any events, especially natural disasters. This paper focuses on the development of a classification model using the messages from the past news websites and tweets during an earthquake disaster event in order to identify the purpose of the message posted to Twitter about the current incidents related to earthquake disaster. A comparison between the Support Vector Machine and Naive Bayes have been evaluated in terms of accuracy. The results show that the support vector machine given highest accuracy than other. However, this approach can be served as useful tools for organization in order to prepare disaster management plans and relief.