Sentimen Analisis Tweet Berbahasa Indonesia Pada Pilkada Serentak 2020 Menggunakan Metode Naive Bayes Berbasis Particle Swarm Optimization

Abstract

Sentiment analysis is a form of text mining, the main focus is analyzing text documents. Tweets in the form of text are divided into two classes, namely positive and negative classes. The algorithm used in this study is naïve bayes based on particle swarm optimization which is used to determine whether there is an increase in the accuracy of the classification results. The dataset used is 1000 and tested using 10 fold cross validation. The classification results obtained from this study produce an accuracy of 81% these results are better than the classification results using naïve bayes without any features selection process with particle swarm optimization an accuracy of 74.14%