ANALISIS SENTIMEN PEMILIHAN CALON PRESIDEN 2024 MENGGUNAKAN METODE NAIVE BAYES

In 2024 Indonesia will hold a democratic party to elect the Indonesian head of state. Every political figure who is nominated as head of state will consider their popularity based on public opinion. Currently, it has had a big impact in building public political opinions, views, sentiments and preferences (ahead of the General Election). Especially on social media, one of which is Facebook. There are several uses for social media, such as meeting new friends, finding out information about sports, economics, tourism and also political matters. One of them is the leader figure of the 2024 presidential candidate, so the author wants to know what information can be taken from public opinion on Facebook social media regarding the leader figure of the 2024 presidential candidate. This problem can be overcome by conducting research in the field of Sentiment Analysis, which is a field of research that focuses on the computational study of opinions, behavior and emotions towards an entity expressed in text form. This research was conducted to find out the results of sentiment analysis regarding the public's response to news about the 2024 presidential candidates and to classify them into three classes, namely positive, negative and neutral using the Naive Bayes method. From 700 comment data on Facebook for each 2024 presidential candidate's fanpage, several duplicate and unimportant data were deleted, resulting in 554 comment data. The accuracy results using the naïve Bayes classifier algorithm were 89.84% which was obtained by using a comparison of 20% test data and 80% training data.

Keywords: Presidential Candidates, Social Media, Facebook, Sentiment Analysis, Naive Bayes