ANALISIS SENTIMEN MASYARAKAT TERHADAP PANDEMI COVID-19 PADA SOSIAL MEDIA MENGGUNAKAN NAÏVE BAYES CLASIFIER

Social media are current communication media widely used by internet users, one of which is Twitter, a site which serves as microblogging provider so its users can share ideas, opinions, or life through short writing called Tweet. The users’ tweets are diverse, consisting data which can be processed into Sentiment Analysis to produce beneficial information to some parties. This study is aimed at creating a sentiment analysis system which can produce data and information in the form of positive, negative, and neutral sentiments. The method used to classify the sentiments was Naïve Bayes Classifier. The inputs of this system were in the form of people’s tweets related to the Covid-19 pandemic. Meanwhile, the outputs were data visualization in the form of positive, negative, and neutral sentiments. In classifying tweets using the Naïve Bayes Classifier method, this People’s Sentiment Analysis System with Naïve Bayes Classifier Algorithm can perform sentiment analysis automatically, resulting in nearly 70% accuracy level. The results of the analysis are depicted in the form of tables and visualization in the form of diagram and wordcloud.