<![CDATA[The growth of social media has made YouTube one of the platforms used by the public to express opinions regarding government policies, including the free electricity program. The large number of comments makes manual analysis difficult; therefore, a text classification method is needed to automatically categorize comments. This study aims to classify YouTube user comments related to the free electricity program using the Naïve Bayes algorithm. The research data were obtained through a crawling process from ten YouTube videos discussing the free electricity policy, resulting in 910 comments, which were reduced to 906 comments after data cleaning. The data processing stages included cleaning, case folding, tokenizing, normalization, stopword removal, stemming, and term weighting using TF-IDF. Furthermore, the data were classified into four categories: Public Discussion and Information, Policy Support and Appreciation, Complaints and Technical Issues, and Non-Electricity. Model evaluation was conducted using a confusion matrix with accuracy, precision, recall, and F1-score metrics. The results showed that the Naïve Bayes algorithm provided fairly good classification performance with an accuracy of 70.9%, precision of 0.62, recall of 0.80, and F1-score of 0.70. The Non-Electricity category achieved the best performance with precision of 0.77, recall of 0.90, and F1-score of 0.83. Based on these findings, the Naïve Bayes method is considered effective for classifying public opinion from social media comment data.]]>
