<![CDATA[Advances in technology and the growth of e-commerce have generated a large amount of user sentiment data, which can be leveraged to improve user experience. Shopee is an e-commerce platform in Indonesia that allows users to buy and sell various products online. Through the Shopee app, users can find a variety of products ranging from clothing, electronics, cosmetics, household needs, to food. However, sentiment data analysis is often constrained by data complexity and the low accuracy of conventional models. This research aims to improve accuracy in Shopee app users by applying Neural Network algorithm. Evaluation of application user sentiment is important for e-commerce companies, such as Shopee, because it provides an overview of user satisfaction and experience with the services offered. In this research, the Neural Network algorithm is used to process sentiment data by optimizing parameters to improve the optimal parameters, including the best K value that can improve model performance. Experimental results show that the best K value is 2, with model accuracy reaching 95.08%. To further measure effectiveness, recall and precision values were calculated for positive and negative categories. The recall result for positive sentiment reached 98.77%, while for negative sentiment it was 93.33%. In addition, the precision for the positive category was 93.73% and for the negative category was 93.33%.]]>
Copyrights © 2025
