Sentiment Analysis of Public Opinion on Twitter about the Implementation of the Merdeka Curriculum Using the Support Vector Machine Algorithm

In 2022 the Head of Standards, Curriculum, Education Assessment (BSKAP) No. 0441HIKR/2022 issued a Decree (SK) regarding Education Units for the Implementation of the Merdeka Curriculum in 140 thousand educational units in Indonesia. The implementation of this new curriculum raises various public opinions on social media Twitter. Therefore, research on the analysis of public sentiment on the implementation of the Merdeka Curriculum was carried out. The tweets taken are tweets with positive and negative sentiments. This study aims to see whether the SVM algorithm is good at carrying out text classification for sentiment analysis of the Merdeka Curriculum by looking at the accuracy value and describing how the public's sentiment on Twitter is about the Merdeka Curriculum. The data used is a dataset of 1,186 tweets, that is 363 positive tweets and 823 negative tweets. The results showed that the accuracy of the Support Vector Machine algorithm with linear, polynomial, and sigmoid kernels was 91.82% and the RBF kernel was 89.88%. As well as seen from the results of the sentiment analysis of implementing the Merdeka Curriculum there were more negative responses, one of which was regarding too many assignments and projects that made students feel tired and stressed.