<![CDATA[This study was motivated by the increasing use of e-commerce in Indonesia, which highlights the importance of analyzing customer reviews as a basis for evaluating product and service quality. This study aims to analyze the sentiment of reviews of Honda motorcycle spare parts at the Ducks Garage store on the Tokopedia platform using the Support Vector Machine (SVM) algorithm. The dataset used consists of 2.537 reviews obtained through web scraping techniques and processed through text preprocessing stages, including data cleaning, normalization, tokenization, stopword removal, and stemming. Sentiment labelling was carried out into three classes, namely positive, negative, and neutral, with lexicon-based and feature weighting using the Term Frequency–Inverse Document Frequency (TF-IDF) method. Data distribution imbalance was handled using the Synthetic Minority Over-Sampling Technique (SMOTE) method. The SVM model was tested using three types of kernels, namely Linear, Polynomial, and Radial Basis Function (RBF). The test results showed that the RBF kernel produced the best performance with an accuracy of 92.79%, followed by the Linear kernel at 89.89% and the Polynomial kernel at 72.57%. The conclusion of this study shows that the application of SVM with SMOTE data balancing is effective in classifying the sentiment of e-commerce product reviews and can be used to support data-driven business decisions based on customer data.]]>
Copyrights © 2026
