<![CDATA[The growth of e-commerce in Indonesia continues to rise with the increasing use of internet and digital technology. Lazada, one of the largest e-commerce platforms in the country, offers a product review system that reflects customer satisfaction. However, the large volume of reviews makes it difficult for sellers to manually evaluate satisfaction levels. This study aims to predict customer satisfaction based on product reviews on Lazada Indonesia using the C4.5 Decision Tree algorithm. The research is quantitative with a data mining approach. Review data was obtained from Kaggle and went through preprocessing, exploration, model building, and testing stages. The attributes used include product name, category, price, number of reviews, and average rating, which were classified into satisfaction levels. The model was evaluated using accuracy, precision, recall, and F1-score. Results show that the C4.5 algorithm can predict customer satisfaction with 77.85% accuracy, 76.74% precision, 83.54% recall, and an F1-score of 80.00%. Product price is the most influential attribute in determining satisfaction. These findings are expected to help sellers understand the key factors influencing customer satisfaction and support improvements in service and product quality on e-commerce platforms.]]>
Copyrights © 2025
