Implementasi Algoritma Naïve Bayes Terhadap Analisis Sentimen Ulasan Aplikasi MyPertamina Pada Google Play Store

Abstract

The application made by PT. Pertamina is an application called MyPertamina. Aside from being a media for transactions, MyPertamina application is also give a rewards and e-vouchers to customer that can be used for transaction benefits at various Pertamina gas stations. Purposes of this research to obtain sentiment data in reviews of the MyPertamina application from the Google Play Store in the form of a text also it looks at an opinion from a user's point of view divided into positive or negative. Data retrieval on MyPertamina application reviews is processed using web scraping techniques using the Google Colab website. The data that has been obtained is then labeled either positive or negative. Data that has been labeled is then cleaned through a preprocessing process which will later be classified by implementing the Naïve Bayes Algorithm. This classification aims to find accuracy, precision, and recall values from reviews of the MyPertamina application data that have been obtained. After classifying, the data was then evaluated and the results obtained were an accuracy value of 77.42%, a precision value of 49.98%, and a recall value of 76.87%.