Analisis Sentimen Masyarakat Terhadap Penerapan Work From Home Berdasarkan Data Twitter Menggunakan Metode Naïve Bayes Classifier Berbasis PSO

At the beginning of 2020 the world was shocked by a virus outbreak which then spread rapidly in various countries, one of which was Indonesia. This virus is known as Covid-19. With a very high spread causing chaos and affecting many sec-tors of people's lives. So that to suppress the spread of Covid-19, the government has made a strategy by establishing a Large-Scale Social Restriction (PSBB) policy. The enactment of this rule aims to suppress the increase in positive cases of Corona Virus Disease 2019 by reducing community mobility. In an effort to strengthen Large-Scale Social Restrictions, the government implemented the Work From Home policy during the emergency period for the spread of Covid-19 which resulted in the emergence of pros and cons to this policy. This has resulted in many people writing their opinions on the Twitter social media platform. So that in this study aims to classify public opinion into positive and negative opinions using sentiment analysis. The data source used comes from Twitter from keywords related to work from home. This study uses the Naïve Bayes method based on Particle Swarm Optimization in building a senti-ment analysis model. The results show that the best accuracy is 81% in the classifica-tion using naive bayes, and 86% in the classification using naive bayes based on PSO