<![CDATA[The rapid development of e-commerce in Indonesia has propelled Tokopedia to become one of the leading marketplaces, offering a diverse range of electronic products from various sellers. In this competitive environment, customer reviews are a crucial factor influencing purchasing decisions and public perception of a store's product and service quality. This research aims to analyze the sentiment of customer reviews for electronic products at the Studio Ponsel store using the Naïve Bayes algorithm. Data was collected through web scraping from Tokopedia, totaling 11,943 reviews. The analysis stages included data cleaning, text pre-processing (normalization, stopword removal, tokenization, and stemming), sentiment labeling (positive, neutral, negative), vectorization with TF-IDF, and model evaluation using accuracy, precision, recall, and F1-score metrics. The evaluation results show that the model achieved an accuracy of 90.27%, with positive sentiment dominating the overall reviews. These findings provide valuable insights for businesses in formulating strategies to improve service, product quality, and marketing effectiveness based on customer data]]>
Copyrights © 2025
