The Sentiment Analysis of Comments on Youtube Channel Beauty Vlogger in Indonesian Language Using Support Vector Machine Method

Siti Chairani Siregar (Department of Informatics, Malikussaleh University)
Rizal Tjuet Adek (Department of Informatics, Malikussaleh University)
Zahratul Fitri (Department of Informatics, Malikussaleh University)

Proceedings of International Conference on Multidisciplinary Engineering (ICOMDEN)
Vol. 2 (2024): Proceedings of International Conference on Multidisciplinary Engineering (ICOMDEN)

YouTube has become a major platform for beauty vloggers to share product reviews, where user comments provide valuable feedback. This research aims to analyze the sentiment in comments from Indonesian speaking users on beauty vlogger channels, focusing on reviews for powder and skincare products, to capture the positive or negative sentiments. The study utilizes the Support Vector Machine (SVM) method for classification and the TF-IDF weighting technique, analyzing 1,000 comments split into 800 training data and 200 testing data. Sentiment classification was performed post-text preprocessing. The results demonstrate a model accuracy of 97%, with a precision of 98% and recall of 96%, indicating that SVM effectively identifies sentiment in user comments. This system provides valuable insights for beauty vloggers to understand product feedback and contributes to the development of similar applications in other industries.