Analisis Sentimen Sunscreen Azarine dengan Naïve Bayes di Toko Aneka Kosmetik Kupang pada Marketplace Shopee

Abstract

Advancements in information and communication technology have changed the way customers shop and share experiences through reviews. Marketplaces like Shopee allow customers to rate products through reviews, making sentiment analysis crucial for understanding consumer perceptions. The Naïve Bayes algorithm is used in this study to analyze 3,504 reviews of the Azarine sunscreen product from Aneka Kosmetik Kupang on Shopee, followed by a text preprocessing process. The dataset is then split into 80% for training and 20% for testing, with reviews categorized into three sentiment classes: positive, negative, and neutral. Evaluation with a Confusion Matrix resulted in an accuracy of 84%, demonstrating the reliability of this algorithm in analyzing customer reviews. The findings of this study offer fresh perspectives for brand owners and potential buyers regarding public perception of the Azarine sunscreen product at Aneka Kosmetik Kupang.