<![CDATA[Sentiment analysis of user reviews on mobile app distribution platforms is a complex and crucial issue, especially with the rapid growth of the number of users and the volume of reviews. This research focuses on the application of Naive Bayes Algorithm to analyze the sentiment of user reviews of WHOOSH app on Google Play Store. Naive Bayes algorithm was chosen due to its efficiency and easy implementation. Using a dataset of 500 cleaned and labeled reviews (positive or negative), the model was trained and achieved 81.25% accuracy. The high precision for the positive class (90%) demonstrates the model's ability to correctly identify positive reviews. Although the recall of the positive class is high (94%), the recall of the negative class still needs to be improved (64%). Overall, the Naive Bayes model is effective for classifying sentiment in WHOOSH user reviews, but needs to improve the accuracy and recall of negative classes.]]>
Copyrights © 2024
