TWITTER SENTIMENT ANALYSIS OF POST NATURAL DISASTERS USING COMPARATIVE CLASSIFICATION ALGORITHM SUPPORT VECTOR MACHINE AND NAÏVE BAYES

Abstract

Natural disasters trigger people, especially Twitter users to provide information or opinions in the form of tweets. The Tweet can be an expression of sadness, concern, or complaint. Processing of data from these tweets will create trends that can be used for information needs such as education, economics, and others. Natural disasters are events that threaten human life caused by nature, including in the form of earthquakes. The method used is the Support Vector Machine and Naive Bayes from the tweet. The data collected is filtered from tweets by deleting duplicate data. In calculating the Natural Disaster sentiment analysis using a comparison of the Support Vector Machine and the Naive Bayes algorithm, the difference in accuracy is 3.07% where the results of the Support Vector Machine are greater than Naive Bayes. The purpose of this research is to analyze sentiment for the distribution of disaster aid that does not flow information due to information & coordination in the field. so as to provide information on the location of natural disasters, natural disaster management, and its presentation to victims that can be shared evenly in an efficient time due to information and natural management so that the distribution of aid is hampered