| Industry 4.0 revolutions where almost all processes using digitalization to ease their business, for example is Internet of Things. Due to the rapid growth of the internet, social networking site has become more popular which usually used for opinion or review sharing about related content. Nowadays, customers are relying mostly on other customer’s reviews. Due to customer’s review can affect public attentions, the company should analyze what people’s think about their company. One of the most popular social networking sites to complain or give some reviews is Twitter. Sentiment analysis is a method to analyze what customer’s think about. Gojek and Grab are the most used applications in Indonesia which provide variety of services especially online transportation service. Both of these companies have very tight competition. In order to improve the service, customer’s opinion on Twitter will be analyzed and classified by Naive Bayes classifier. Data was collected by quota sampling up to 1000 tweets for each company. The classifier has performed well with the average of accuracy, sensitivity, precision, specificity was 90%.Most of frequent words are visualized on word cloud and find related word associate by R Studio. All negative sentiments are visualized on fishbone diagram to find the solutions. |