Natural disaster topic selection using decision tree classification

Abstract

Social media like Twitter used as a reference for popular and trusted news today. This opportunity could be useful manner to inform the natural disasters that occur in certain condition related to their locations. This study aims to proposed model to select the topic related to the natural disasters that occur in Indonesia based on Twitter user timeline. The selected topic will use as a reference to ascertain events, conditions of victims, and position of disaster locations. In this study, the data taken from the official Twitter account of @BNPB_Indonesia. The data extracted into two main classes, and grouped based on types of disasters, victims, and disaster locations or safe point position using decision tree algorithm.