Analisis Sentimen Masyarakat terhadap Kebijakan Program Makan Bergizi Gratis (MBG) di X Menggunakan Metode Support Vector Machine (SVM)

The Free Nutritious Meal (MBG) policy has elicited diverse responses on social media platform X, necessitating sentiment analysis to objectively understand public opinion. This study employs Support Vector Machine (SVM) with TF-IDF weighting to classify 3,796 opinion data into positive, negative, and neutral sentiments. Results show the SVM model achieves an accuracy of 76.05%. Public perception is predominantly positive (1,431 data), followed by neutral (1,258), and negative (1,107). Word Cloud analysis indicates optimism based on hopes for improved student nutrition, while negative opinions highlight concerns about food safety and budget transparency. In conclusion, MBG enjoys strong public support, but the government should prioritize food safety and transparent financial management to reduce public unrest. Identifying negative sentiments provides a basis for critical evaluation and future policy improvements.