Klasifikasi Sentimen Ulasan Coffee Shop Di Ponorogo Berbasis Naïve Bayes

Indonesia

Abstract

This study aims to analyze customer sentiment toward three coffee shops with the highest number of reviews in Ponorogo, namely Gayeng Billiard & Naïve, Kedai Sor Sawo, and Origin Lab Coffee. Customer review data were collected through web scraping from Google Maps and processed using the CRISP-DM framework, which includes text cleaning, normalization, tokenization, and feature extraction using the TF-IDF method. The Naïve Bayes algorithm was employed as the primary classification model to categorize sentiment into positive, negative, and neutral classes. Additionally, this research compares sentiment tendencies across the three coffee shops to identify differences in customer perceptions based on the available reviews. The analysis results show that most reviews exhibit positive sentiment, with Variations in customer satisfaction across the three coffee shops. The Naïve Bayes model achieved a satisfactory level of accuracy in classifying sentiments. These findings are expected to provide practical insights for coffee shop owners in understanding customer needs, evaluating service quality, and formulating strategies to enhance both operational performance and customer experience.