Social media are used by people as the platform to express their opinions and the conditions that happen around them. During the ministerial election for 2019-2024 Cabinet of Indonesia Maju, the choice of ministers of President Indonesia Jokowi always present in the discussed topic on a social media. The most discussed topic is the appointment of Nadiem Makarim as the Minister of Education and Culture. In Indonesia, the ministers have to show their performance and capability. If they failed to do so, they would be reshuffled by the President. One of the sources of information required by President to be able of evaluating the performance of his ministers is through the feedback from citizens. Sentiment analysis is a computation study of opinions, attitudes, and emotions of people toward entities, individuals, issues, events, or topics. Sentiment Analysis The targets of sentiment analysis are to discover arguments, identify the expressed sentiments, and then classify their polarity into positive, negative, or neutral categories. In principle, the classification methods on sentiment analysis can be performed through machine learning approach. Some classification methods of machine learning such as Decision Tree, K-NN, Naïve Bayes, and Random Forest are often used to acquire the best result. Next, several stages were done, including the data collection by crawling Twitter's data via API with “Nadiem Makarim” as the keyword, pre-processing, classification and evaluating the classification performance. Naïve Bayes acquired the best result with 99% accuracy, 94% precision, and 99% Recall. It can be concluded that Naïve Bayes is the best classifier to be used on the dataset of Indonesian-Language social media because it can provide the most accurate and appropriate prediction.