ANALISIS SENTIMEN ULASAN PRODUK SKINCARE DI SHOPEE UNTUK MENINGKATKAN KUALITAS PRODUK MENGGUNAKAN METODE SUPPORT VECTOR MACHINE

Abstract

This research discusses sentiment analysis of skincare product reviews on Shopee using the Support Vector Machine (SVM) algorithm. The research stages include review data collection, data pre-processing (cleaning, tokenization, stemming, and stopword removal), feature weighting using Term Frequency-Inverse Document Frequency (TF-IDF), and SVM model training. The evaluation results show that the SVM algorithm achieved an accuracy of 93.79%. For the negative sentiment category, precision reached 100% with recall of 18% and F1-score of 31%, while for the positive sentiment category, precision was 94%, recall 100%, and F1-score 97%. The weighted average values for precision, recall, and F1-score are 94%, 94%, and 92%, respectively. The results of this study show that the SVM algorithm is able to provide good performance in sentiment analysis, especially in the dominant category in the dataset, although it requires improvement in handling data imbalance. This research is expected to help decision makers in understanding consumer preferences and improving product quality.