Analyzing public sentiment on youtube comments regarding the free lunch policy using the Support Vector Machine (SVM) algorithm

Abstract

The advancement of information technology and social media has reshaped how individuals express their opinions on public policies. YouTube has emerged as a major platform where public sentiment is openly shared, including reactions to the government’s Free Lunch Program for elementary school students. This study aims to analyze public sentiment toward the policy using the Support Vector Machine (SVM) algorithm with both linear and Radial Basis Function (RBF) kernels. A total of 1,883 YouTube comments were collected and manually labeled into three sentiment categories: positive, negative, and neutral. The preprocessing steps included cleansing, case folding, normalization, tokenization, stopword removal, and stemming, followed by TF-IDF transformation. Model performance was evaluated using accuracy, precision, recall, and F1-score metrics, and validated using 10-Fold Cross Validation to ensure result consistency. The findings indicate that the SVM model with RBF kernel and 10-fold cross-validation achieved the highest accuracy at 81.46%. However, the linear kernel model provided a more balanced performance with superior precision, recall, and F1-score. These results highlight the importance of choosing the right kernel and validation strategy in developing sentiment analysis models, especially when dealing with imbalanced social media data.