Customer Review Sentiment Classification of Belikopi Products Using the Support Vector Machine (SVM) Method

Abstract

The rapid development of information technology and the widespread use of social media have significantly changed the way customers express their opinions and experiences regarding products and services. Platforms such as Instagram, TikTok, and Google Maps have become important sources of customer feedback that can be analyzed to understand public perception and customer satisfaction. Sentiment analysis is one of the text mining techniques that can automatically classify opinions into positive, neutral, and negative sentiments, enabling businesses to make informed decisions based on customer feedback. This study aims to analyze customer sentiment toward Belikopi products using the Support Vector Machine (SVM) classification algorithm. A total of 4,636 customer reviews and comments were collected from Instagram, TikTok, and Google Maps and manually labeled into three sentiment categories: positive, neutral, and negative. Before the classification process, the dataset underwent several preprocessing stages, including case folding, cleaning, tokenizing, stopword removal, and stemming to improve the quality of textual data. Furthermore, the Term Frequency–Inverse Document Frequency (TF-IDF) method was employed to convert text into numerical feature vectors suitable for machine learning classification. The dataset was divided into 80% training data and 20% testing data using a stratified sampling approach to maintain the distribution of sentiment classes. The experimental results showed that the SVM model achieved an accuracy of 93.34%, demonstrating its capability to classify customer sentiment with high performance. The findings indicate that the proposed approach is effective in identifying customer perceptions of Belikopi products and can provide valuable insights for evaluating customer satisfaction, improving product quality, and supporting strategic business decision-making.