<![CDATA[The free nutritious meal program, has sparked various reactions from the public, particularly on social media. This study aims to determine how netizens feel about the program through data analysis from social media platform X (formerly Twitter). Data was collected using web scraping techniques with the keyword “free nutritious meals,” then processed through text cleaning and automatic labeling stages using the Indonesian language version of the RoBERTa model. With this approach, each tweet was efficiently and accurately classified into positive or negative sentiment. The labeled data was then analyzed using two classification algorithms: Support Vector Machine (SVM) and Naïve Bayes. Test results showed that SVM performed better, with an average accuracy of 0.8367 and a deviation of 0.0117. Meanwhile, Naïve Bayes recorded an accuracy of 0.7716 with a deviation of 0.0101. Visualization through WordCloud also shows the dominant words in each sentiment. Words such as “support,” “healthy,” and ‘grow’ appear frequently in positive sentiments, while words such as “budget,” “poison,” and “cost” dominate negative sentiments. These findings illustrate significant public support for the program, but also concerns regarding its implementation and funding. The results of this analysis are expected to provide input for policymakers in understanding public opinion more objectively]]>
Copyrights © 2025
