Analisis Sentimen Review Pelanggan Pada Layanan Indihome Menggunakan Algoritma Naïve Bayes Classifier

Abstract

Abstract: PT. Telkom Indonesia as a customer-oriented company certainly must understand and know how important customer assessment of services, agencies or organizations. at PT. Telkom Indonesia has a problem that is Telkom Indonesia does not know about the opinions or customer reviews of IndiHome services on websites and social media. At present Telkom Indonesia only knows customer reviews or opinions through complaints directly or via Customer Service (CS) telephone. Based on the existing problems, the solution is obtained in the form of an application that can display customer reviews whether positive or negative. Customer reviews can be processed words / sentences using the text mining method and words / sentences can be classified as positive and negative classes using the naïve Bayes method. From the results of the sentiment analysis research it can be concluded that the sentiment analysis in this study can display a visualization of the results of the sentiment analysis classification in the form of pie charts and produce a sentiment analysis classification as evidenced by 100% function testing, testing calculations with different values with the same classification, and 100% accuracy testing with predictive value / actual value is True Negative (TN). keywords: Sentiment Analysis, Telkom Indonesia, Naïve Bayes, Text Mining.