<![CDATA[This study aims to analyze user sentiment toward mobile applications based on reviews collected from Google Play Store by applying the Naive Bayes classification model. User reviews represent an important source of information that reflects user experiences, satisfaction levels, and perceived application quality. However, the large volume and unstructured nature of textual reviews make manual analysis inefficient and subjective. Therefore, this research adopts a quantitative approach using text classification based on machine learning to automatically categorize user reviews into positive, negative, and neutral sentiment classes. The research process consists of data collection, text preprocessing, feature extraction, sentiment classification using Naive Bayes, and model performance evaluation. Text preprocessing includes case folding, tokenizing, stopword removal, and stemming to improve data quality. Model performance is evaluated using accuracy, precision, recall, and F1-score metrics. The results show that positive sentiment dominates user reviews, indicating that the application is generally well received by users, although negative and neutral sentiments remain present and highlight areas that require improvement. The evaluation results demonstrate that the Naive Bayes model achieves reliable performance in classifying sentiment, with balanced evaluation metrics that indicate stable classification capability. These findings confirm that Naive Bayes remains an effective and efficient method for sentiment analysis of application reviews. This study contributes theoretically to sentiment analysis research and practically provides insights that can support application developers in evaluating user feedback and improving application quality.]]>
Copyrights © 2026
