CHARACTERIZATION OF VALUABLE INFORMATION FROM SOCIAL MEDIA NETWORKS DURING NATURAL DISASTERS

Twitter, a microblog site, have emerged as a new source for detecting and monitoring disaster events specifically earthquakes. The data streamed in Twitter can be used to pull actionable data for emergency response and relief operation. However, no effort has been made to classify data in conversational and informative form which has been used as a common reference for the decision-makers to streamline priorities and activities during a disaster. This paper makes an initial effort in classifying tweets by examining more than 10,000 tweets generated using the hashtags, #Lindol, #EarthquakePH, #Mindanao and the word ‘Lindol' as a mention. The results are generated using Rapidminer software and deemed as necessary and useful for the disaster management unit. Information generated from the classification can also be used by the social science research communities to study various aspects of preparedness, response, impact and recovery of disaster.