Perbandingan Metode Naïve Bayes Classifier dengan Metode Random Forest pada Prediksi Rating Review Drama Korea

Abstract

Korean dramas have very many fans and are spread in various countries. This study aims to determine whether the korean drama is classified as Bagus, Tidak Bagus, or Cukup Bagus and compares two methods, namely the naïve bayes classifier method and the random forest method in predicting korean drama review ratings. This study shows that the naïve bayes classifier and random forest methods are capable of predicting korean drama review ratings. In the prediction review, the random forest method obtained an accuracy value of 89%, while the naïve bayes classifier method obtained an accuracy value of 86%. In rating predictions, the random forest method obtains an accuracy value of 41%, while the naïve bayes classifier method obtains an accuracy value of 40%. The conclusion of this study is that the random forest method is superior and accurate in predicting Korean drama review ratings.