KLASIFIKASI SENTIMEN MASYARAKAT PENGGUNA TWITTER MENGGUNAKAN METODE NAÏVE BAYES PADA KASUS PEMERINTAHAN DAERAH

Abstract

Twitter is a microblogging site that allows its users to write about various opinions, comments, and news that discuss current issues. Many users post their opinion on a product or service they use. It can be used as a source of data to assess sentiment on Twitter. One method of automatic emotional grouping can be used, one of which is using Naïve Bayes. The purpose of this research is to build a system that is able to automatically classify the emotions of each tweet, and to determine the accuracy of the grouping.The process starts from preprocessing, there are several processes, namely tokenizing, stopword, stemming, word weighting, and normalization, which can then be processed using Naïve Bayes. Naïve Baye Process The creation of a sentiment analysis system using the Naïve Bayes method has proven that the algorithm can analyze sentiments automatically, with an accuracy rate of 95%. The results of this visualization can be used by the Government to determine policies to be taken in the future. After that perform accuracy calculations using confusion matrix.

Keywords:Tweet, Naïve Bayes, governmentMaksimum