ABSTRACT
Nowadays, microblogging sites have become a very popular communication tool
among internet users. Microblogging is a social media service that allows users to
publish short messages in the form of opinions, comments, news in limited
characters (less than 200 characters). One of these microblogging services, which
is quite a lot of users in almost all circles is Twitter. Based on the tweets that were
trending in February 2024 after the Indonesian presidential general election, there
were many tweets with opinions regarding Prabowo Subianto, the 8th president of
Indonesia, including positive, negative or neutral opinions. Based on this, it is
necessary to create a sentiment analysis system to make it easier to analyze and at
the same time classify the responses of the community of Twitter users based on
their class, whether the tweet is positive, negative, or neutral. In order to do this
classification, we need an algorithm that can perform classification. One of the
algorithms that can be used is the Naïve Bayes classifier, an algorithm for
classification by calculating probability/possibility values. This method can be used
to classify tweets. The data set in the form of tweets that have been collected will
be used as training data and 246 test data with details of 96 training data and 150
test data. The result of accuracy using these data is 84%.
Keyword: tweet, naïve bayes classifier, classification.