ANALYSIS OF TWITTER SENTIMENT CLASSIFICATION ON TELECOMMUNICATION PROVIDER SERVICE PERFORMANCE USING NAÏVE BAYES

Abstract

Background - Telecommunication users in Indonesia continue to grow rapidly every year. Due to the increasing
needs of users for communication and data. This results in competition between telecommunication providers to
attract or even retain customers. Several studies have stated that the Customer Opinion factor can indicate the
degree of quality of the provider. Various user opinions about telecommunication providers can be mined through
social media, one of which is Twitter.
Purpose - Twitter is one of the microblogging that produces data that floods its users. The benefits that can be
taken from the data are to classify data for analysis.
methodology - One method for classification is Naïve Bayes can handle text. The text used in this study is Twitter
user comments. The Naïve Bayes method was chosen because of its ability to handle large text data with
adequate accuracy and efficient computing processes.
Findings - The results of this analysis are expected to provide insight for telecommunication providers in
improving the quality of their services according to user needs and expectations. This study categorizes the
sentiment opinions of telecommunication provider users.
Originality - The expected Value of this research are that telecommunications providers can evaluate
performance and services to achieve customer satisfaction from various complaints faced, as well as build more
effective communication strategies.