Analisis Sentimen terhadap Perpanjangan Masa Jabatan Presiden Indonesia Menggunakan Naïve Bayes

Twitter social media is often used as a medium to express opinions on the government in Indonesia. There are also many controversial things on Twitter social media against the government, as is the case today there is a controversy wherethere is a proposal to make the term of office of the President in Indonesia three periods, which previously could only serve up to two terms or ten years. for one President. Public opinion or sentiment which in Twitter's terms is commonly referred to as "shrink" can be in the form of negative or positive opinions. However, the amount of data is quite large so it takes a method that can be used to make it happen, namely by using sentiment analysis. Sentiment analysis can be used as a solution to process these opinions using the Naïve Bayes Classifier algorithm. The results of the nave Bayes technique require evaluation to determine the best model. The test is carried out with three models, namely 70:30, 80:20, and 90:10 which are then evaluated using a confusion matrix. Based on the comparison of the evaluation test results, the best scenario for the Naïve Bayes classification model is in the first scenario or model (90% training data and 10% testing data) with an accuracy value of 95%, a precision value of 97%, a recall value of 96%, and the f-measure value is 96%.