<![CDATA[Public policy discussions increasingly take place on social media, where public opinion is shaped not only by message content but also by patterns of user interaction. This study analyzes online conversations related to Indonesia’s Free Nutritious Meal (MBG) policy on platform X by integrating Social Network Analysis (SNA) and sentiment analysis. The dataset consists of 3,459 tweets collected between January 8 and February 11, 2025. Communication networks were constructed based on reply and mention relationships to identify interaction patterns and influential accounts. Sentiment analysis was conducted using a Natural Language Processing approach, with initial labeling based on BERT and further classification using a Support Vector Machine (SVM). Model performance was evaluated using accuracy, achieving a score of 91.78%. The findings reveal that MBG discussions form a relatively sparse yet highly centralized network dominated by a small number of accounts. Most tweets express neutral sentiment, while temporal analysis indicates a significant spike in activity on February 10, 2025. This study demonstrates that integrating network and sentiment analysis provides a more comprehensive understanding of how public opinion evolves in digital environments.]]>
Copyrights © 2026
