Analisis Sentimen Terhadap Operator Seluler Telkomsel Menggunakan Algoritma Naïve Bayes

Abstract

The use of social media in the era of globalization is very necessary for all people including Telkomsel cellular operators, Telkomsel uses Twitter as a promotional media. However, the use of Twitter as a medium for promotion of Telkomsel received various compliments of input and criticism and in this case the authors conducted research to analyze Telkomsel user sentiment. And in this study the authors used the Naïve Bayes algorithm to classify sentiments and look for preference values. From the results of tests conducted using cross validation techniques and accuracy measurements using confusion matrix with 10 times the best accuracy testing obtained was 85.33 % and the positive response obtained from the calculation of preference value was 37.03 %. It can be concluded that the Naïve Bayes algorithm can be used to classify quite well and be able to measure user responses.

Keywords : Twitter, Sentiment Analysis, Naïve Bayes Classifier, Cross Validation, Preference Value