Analisis Sentimen Program Mbkm Pada Media Sosial Twitter Menggunakan KNN Dan SMOTE

Abstract

The Merdeka Belajar-Kampus Merdeka (MBKM) program is a relatively new program implemented in Indonesia since February 2020. Like a new program, the implementation of the MBKM program is also followed by various pro and con attitudes. Therefore, a sentiment analysis technique is needed to determine the public opinion towards the MBKM program. The purpose of this study is to determine the performance of the KNN method in performing sentiment classification optimized by the SMOTE method in overcoming the problem of unbalanced data and to determine the tendency of public sentiment towards the implementation of the MBKM program. Based on the research results, the KNN method optimized with the SMOTE method is proven to improve classification performance. From initially producing an accuracy value of 76.13%, precision of 76.03%, recall of 76.13% and f1-score of 76.01% there was an increase in accuracy value to 76.13%, precision to 76.03%, recall to 76.13%, and f1-score to 76.01%. In this study, it was found that community responses tended to be neutral towards the MBKM program. The community feels that the MBKM program is a program that can increase student experience. However, there are still program systems that are considered complicated and need to be evaluated.