COMPARISON OF SENTIMENT ANALYSIS OF SUPPORT VECTOR MACHINE AND NAÏVE BAYES ALGORITHMS ON PUBLIC RESPONSE ABOUT ONLINE LEARNING DURING THE COVID-19 PANDEMIC

Abstract

At the beginning of 2020, COVID-19 began to spread throughout the world, including Indonesia. The government continues to look for ways to prevent the chain from spreading, one of which is by implementing online learning. The background of this research is to use twitter to find out the response and public sentiment about online learning during the covid-19 pandemic. The purpose of this research is to find out public opinion about the application of online learning and also to compare the performance level of support vector machine and naïve Bayes algorithms. In conducting this research, the type of research used is qualitative research in order to be able to understand well what kind of phenomena experienced by the research subjects. The best sentiment analysis results are obtained by comparing two classification algorithms, support vector machine and naïve Bayes. Testing based on k-fold cross validation aims to obtain accuracy, precision, and recall values. The best algorithm will produce the right output with a higher test score.