Sentiment Analysis of Reviews on Virgin America Airlines Using the Decision Tree Method

Abstract

This sentiment analysis aims to evaluate public opinion on airline services by analyzing review data collected from passenger reviews. The methods used includetext processing techniques and statistical analysis to identify and classify sentiments in reviews as positive, negative or neutral. The purpose of this study is to determine the accuracy and user reviews about airlines. This research uses the Decision Tree method with the stages of literature study, data collection, preprocessing, decision tree modeling and evaluation. The results of the Decision Tree method state that the accuracy rate in this method is 52.18%. In class precision for pred. negative 0.00%, for pred positive 73.68% and pred.negative 51.20%. In class recall neutral 0.00%, class recall positive 14.51% and class recall negative 98.85%.