Analisis Sentimen di Youtube Terhadap Kenaikan UKT Menggunakan Metode Support Vector Machine

Abstract

Students and the general public usually have different responses to the increase in Single Tuition Fees (UKT) at universities. Protests, complaints, and support for this increase may be expressed through various social media platforms, such as YouTube. Using the Support Vector Machine (SVM) method, this study analyzes comments on the YouTube platform related to the increase in UKT. Comment data is divided into three categories: positive, negative, and neutral. The evaluation results show that the SVM model achieves an accuracy of 0.88; it also demonstrates good ability to recognize negative sentiment with a precision of 0.83, recall of 0.90, and f1-score of 0.86. For neutral sentiment, the model shows a precision of 0.86, recall of 0.75, and f1-score of 0.80. Nevertheless, the SVM model achieves perfect scores for precision, recall, and f1-score of 1.00 for positive sentiment. Although the SVM model has proven effective in analyzing sentiment towards the increase in UKT on YouTube, further improvements are needed to enhance accuracy in identifying neutral sentiment.