PENERAPAN ALGORITMA NAIVE BAYES CLASSIFIER UNTUK ANALISIS SENTIMEN KOMENTAR TWITTER PROYEK PEMBAGUNAN IKN

Abstract

Recently, the relocation of Indonesia's capital city has become a hot topic of discussion among the public. Various opinions emerged regarding this mega project proposed by President Joko Widodo. Especially on social media, Twitter has become one of the most popular platforms in Indonesia as a forum for expressing people's opinions in public. In the context of the development of IKN Nusantara, researchers conducted an analysis of Twitter users' comments on President Joko Widodo's official account. Using the Naïve Bayes method with a dataset containing 220 comments consisting of 116 negative comments, 70 positive comments and 34 neutral comments. In this research, researchers developed a Python-based machine learning program. The analysis results show respective values of Precision 62%, Recall 66%, and f1-Score 63% with an accuracy level of 66%. In testing using 20% of the data, the program successfully predicted 20 negative comments, 8 neutral comments, and 16 positive comments.