<![CDATA[The rapid development of the e-commerce sector in Indonesia has made customer feedback a very important source of information in assessing the quality of goods and services. However, with so many reviews available, the manual assessment process often becomes complicated. The purpose of this study is to analyze customer sentiment towards PedagangAksesoris store on the Shopee platform using the Naïve Bayes Classifier method to identify positive and negative opinions that can help improve customer satisfaction. The data for this study was collected through web scraping of Shopee user reviews, followed by a preprocessing stage that included cleaning, filtering, removing affixes, and separating words. The data was then divided into training data and testing data to train and test the model. The Naïve Bayes method was applied by calculating word probabilities using Laplace smoothing, while model performance was evaluated using a Confusion Matrix through the RapidMiner application. The results of this study show that the Naïve Bayes model can classify customer reviews with a high degree of accuracy, with precision reaching 100% for the negative category and 80% for the positive category, as well as recall of 87.5% and 100%. These findings confirm that the Naïve Bayes method is an effective and efficient way to perform text-based sentiment analysis on reviews in e-commerce. The results of this sentiment analysis can be used as a basis for strategic decision-making by businesses to improve product quality, services, and customer satisfaction. ]]>
