Perbandingan Algoritma SVM dan Naïve Bayes Berbasis SMOTE dalam Analisis Sentimen Komentar Tiktok pada Produk Skincare

Liem, Steven
Setiawan, Thomas
Pribadi , M. Rizky

Applied Information Technology and Computer Science (AICOMS)
Vol 3 No 1 (2024)

This research compares the performance of the Support Vector Machine (SVM) and Naïve Bayes algorithms in sentiment analysis of TikTok comments about skincare products, using the Synthetic Minority Over-sampling Technique (SMOTE) to address data imbalance. The evaluation results indicate that SVM outperforms Naïve Bayes, achieving an accuracy of 59.43% compared to 47.65%. Additionally, SVM excels in the F1 Score metric (60.37% versus 54.74%), although Naïve Bayes demonstrates slightly higher precision (67.96% compared to 62.76%). Therefore, SVM proves to be more effective in classifying sentiment comments, making it the recommended algorithm for sentiment analysis tasks in the skincare product domain on TikTok.