Sentiment Analysis of Customer Satisfaction on Transportation Network Company Using Naive Bayes Classifier

Abstract

The development of smartphones with a geopositioning system (GPS) embedded on it, paves a way to the online transportation service application. Since then, online transportation platforms such as Gojek and Grab gain its popularity. Transportation network companies such as Gojek and Grab use social media as advertisement media and source of user reviews. Public communicate their opinion about their experience of using an online transportation service through social media. Sentiment analysis helps to extract people’s opinions, sentiments, evaluations, and emotions concerning user experience of certain services. This research considers sentiment in level which is positive, neutral and negative with uses twitter data as the data source. A preprocessing task ensures the quality of the data, cleansing, filtering, tokenizing and stemming were performed before the classification task. From the evaluation result, classification using Bayes Naive algorithm succeeded get accuracy of 72.33% with average recall and precision of 73.95% and 73.24%.