The educational curriculum is a device or system of plans and arrangements regarding learning materials on teaching and learning activities. If the curriculum does not meet the requirements or facilitate teaching and learning activities, then the curriculum cannot be said to be good. The aim of this research is to filter and analyze sentiment from public opinion towards the newest curriculum in Indonesia, namely the independent curriculum, which will be made into the national curriculum in the upcoming 2024. The dataset used is tweets from Twitter as many as 667 lines of tweets labeled as positive and negative categories. The labeling process is done automatically using the VADER sentiment library. In sentiment analysis, one of the classification methods that is quite good in sentiment classification is Naive Bayes and K-Nearest Neighbor (KNN). The Naive Bayes method is fairly good in the data classification process that can study the training data provided to it properly, while KNN is a simple method that is quite easy to understand and is often used in the classification process which produces quite good accuracy compared to other methods. The stages in conducting sentiment analysis in this study are data collection, data preprocessing, labeling or annotation of data, data visualization, classification and evaluation.In addition, the results of sentiment tend to be negative, so it can be concluded that the independent curriculum which will be made into the national curriculum has not been well received by the public, so this can be taken into consideration for the Indonesian government to make the independent curriculum a national curriculum.

Keywords : Sentiment Analysis, Independent Curriculum, Naive Bayes, KNN, VADER, Natural Language Processing.