Analisis Sentimen Kinerja Humas Polda Malut Pada Twitter Menggunakan Metode Naive Bayes Classifier

Abstract

   The current public conversation is Police Performance because the Community gives their opinion about POLICE performance through social media, namely Twitter. There are several methods used to carry out sentiment analysis, one of which is the Naïve Bayes Classifier method. Based on the results of testing the Naive Classifier algorithm that has been carried out, several things have been produced, among others, based on the results of analysis from research that almost the majority of people have a neutral view of the Public Relations performance of the Polda , which is proven by the analysis of this study. In this study, the Naïve Bayes Classifier algorithm produces an accuracy value of 69%, the level of accuracy of sentiment analysis on Twitter social media with the keywords divhumas_polda_malut, poli_malut in research the precision value is 68%, the recall value is 71%, F-measure value 68%.