<![CDATA[The Free Nutritious Meals Program (MBG) is a government policy initiative aimed at supporting nutritional fulfillment and human resource development. However, its implementation has generated diverse public responses, making a measurable analysis necessary to identify trends in public opinion. This study aims to classify the sentiment of Youtube users comments toward the MBG Program using the Support Vector Machine (SVM) algorithm. The study employed a quantitative descriptive-analytical approach using comments from a video published by the Ngomongin Uang Youtube channel entitled “Program Makan Bergizi Gratis (MBG): Solusi Ekonomi atau Beban Negara?”. A total of 1.842 comments were collected through the Youtube Data API v3, of which 1,821 comments were retained after the data-cleaning process. The analytical stages included case folding, data cleaning, tokenization, stopword removal, stemming, lexicon-based labeling, TF-IDF weighting, and classification using LinearSVC. The results showed that 1.471 comments were classified as negative (80,78%), 222 comments as positive (12.19%), and 128 comments as neutral (7,03%). The model correctly predicted 300 out of 365 test data, achieving an overall accuracy of 82.19%. These findings indicate the dominance of critical responses and confirm the usefulness of SVM for mapping public opinion through Youtube comments]]>
Copyrights © 2026
