PERBANDINGAN METODE NAÏVE BAYES, DECISION TREE, DAN KNN DALAM ANALISIS SENTIMEN APLIKASI GOJEK DI PLAYSTORE

Maretta, Aulia Pinkan
Anadia, Qothrunnada Wafi
Sasmita, Ruth Mei
Epriyanti, Nadia
Rizkyllah, Anabel Fiorenza
Mariska, Inneke Via
Tania, Ken Ditha
Meiriza, Allsela

Zonasi: Jurnal Sistem Informasi
Vol. 7 No. 2 (2025): Publikasi artikel ZONAsi: Jurnal Sistem Informasi Periode Mei 2025

Sentiment analysis on user evaluation of Gojek application services on Play Store is important to understand user opinions on the services provided. This study compares three machine learning methods, namely Naïve Bayes, Decision Tree, and K-Nearest Neighbors (KNN) when categorizing user sentiment on Google Play Store as positive, negative, or neutral. The data processed comes from the Gojek user review dataset obtained from Kaggle. The analysis process involves data preprocessing (cleaning, stopword removal, tokenization, and split data), data transformation, and implementation of classification algorithms. The evaluation was carried out using accuracy, precision, recall, and F1-score metrics. The results of the study prove that Naïve Bayes has the best performance with an accuracy of 89%, followed by KNN (86%) and Decision Tree (84%). This study provides good insight for application developers in choosing the best method to understand user opinions and improve service quality.