ANALISIS SENTIMEN PENERAPAN PPKM PADA TWITTER MENGGUNAKAN NAIVE BAYES CLASSIFIER DENGAN SELEKSI FITUR CHI-SQUARE

Abstract

Dissemination of information related to the implementation of PPKM takes place very quickly, especially on social media networks. Positive and negative news certainly has an impact on public opinion or sentiment on the implementation of PPKM. In this study, the data used is public opinion on Twitter social media with the keyword "PPKM" from November 2, 2021 to November 8, 2021 and obtained data as many as 12,616 tweets which then deleted duplicate data to become 6,465 data. Data classification was performed using Naïve Bayes with Chi-Square feature selection and the data were classified into positive and negative classes. The results of the classification performance using Nave Bayes with Chi-Square feature selection obtained an accuracy of 80% which means that the Nave Bayes classification model with Chi-Square feature selection is quite effective in classifying public opinion on the implementation of PPKM. Keywords: PPKM, Twitter, Sentyment Analysis, Naïve Bayes Classifier, Featured Selection Chi-Square.