<![CDATA[The advancement of information technology has influenced various aspects of life, including transportation. The Access by KAI application provides digital train ticket booking services. With millions of users, analyzing the level of satisfaction through reviews on the Google Play Store is important to improve service quality. This study aims to classify user reviews using the Naïve Bayes algorithm to determine the level of satisfaction, group reviews based on certain categories, and evaluate the accuracy of the classification results. The study uses the Naive Bayes method to classify text where review data collection is carried out first through a scraping process from the Google Play Store with a total of 1000 reviews. Data is analyzed through pre-processing stages such as cleaning, case folding, tokenization, normalization, stopwords, steamming and sentiment labeling using InSetLexicon. Furthermore, reviews are grouped by category using the K-Means clustering method to group data into three categories, namely Features, Services, and Systems, to improve classification accuracy followed by classification using Naïve Bayes. Evaluation is carried out using a confusion matrix to measure accuracy, precision, recall, and F1-score. The classification results show that in the Feature category, precision is 79%, recall 99%, and F1-score 88%. In the Service category, precision reaches 100%, recall 56%, and F1-score 72%. For the System category, precision is 94%, recall 68%, and F1-score 79%. Overall, the model achieves an accuracy of 83%. The benefits of this study are to provide a deep understanding of user needs and become a reference for developers to improve the Access by KAI application service.]]>
Copyrights © 2025
