<![CDATA[Skincare is considered a crucial need in maintaining and caring for skin health, given its role not only in cosmetics but also in contributing to the well-being and protection of the skin from external factors. Sentiment analysis of skincare reviews on social media helps understand consumer perspectives, guiding skincare manufacturers in product improvement. The comparison of polynomial and RBF kernels in SVM is relevant to enhance sentiment analysis of skincare in Indonesia, ensuring the model's accuracy in classifying product sentiments. The dataset used consists of 2168 data obtained through social media scraping. After obtaining the data, text preprocessing processes such as case folding, cleaning, tokenization, stemming, and data labeling were performed. The data was divided into an 80:20 ratio for comparison, with 1734 training data and 434 testing data. The accuracy results using the SVM method with RBF and Polynomial kernels were obtained, with the highest accuracy found in the RBF kernel at 86,17%, and the polynomial kernel achieving an accuracy result of 84,56%.]]>
Copyrights © 2024
