Analisis Sentimen Grab Indonesia pada Ulasan Google Play Store menggunakan Algoritma Naïve Bayes dan Support Vector Machine

Abstract

This study uses the Naïve Bayes and Support Vector Machine (SVM) algorithms to analyze the sentiment of user reviews on the Grab Indonesia app on the Google Play Store.  Web scraping was used to gather the review data, which was then processed through a number of stages, such as tokenization, letter modification, the elimination of unnecessary words, and weighting using the TF-IDF approach.  The findings of the investigation demonstrate that SVM performs better in classifying positive and negative sentiments and has a greater accuracy (93%) than Naïve Bayes (92%).  But in terms of computational efficiency, Naïve Bayes continues to lead the field.  This study sheds light on how well both algorithms do sentiment analysis on Indonesian mobile apps