The democratic system is highly respected in Indonesia. The Indonesian state government organizes its government in a democratic manner, namely, it is run by, for, and with the consent of the people. Democracy is held through general elections to occupy leadership and power seats, followed by political parties. The high enthusiasm of Twitter social media users with the electability of presidential candidates proposed by political parties in the 2024 general election and the existence of a survey of potential presidents from national institutions, as well as mass and social media coverage of presidential candidates suggested by the political parties, generates opinions, attitudes, and emotions among people from all walks of life through tweets. The availability of abundant data on Twitter and other social media can provide useful information. Tweet data is obtained by crawling data using the help of the Python library, namely snscrape. The sentiment analysis uses a mixed method, namely by using machine learning and Lexicon Based, through the process of fine-grained sentiment analysis using the technique, namely, knowing the level of opinion polarity by grouping netizen responses and opinions into three parts: positive, neutral, and negative, with the help of machine learning and natural language processing. The results of the study were carried out by experimenting with four scenarios by dividing test and training data by 60:40, 70:30, 80:20, and 90:10. Measurement of the accuracy value results in the classification of the Nave Bayes Classifier Algorithm as 68%, 67%, 70%, and 71%. From the tweet data, it is clear that positive sentiment is dominant on the research topic.