<![CDATA[The Free Lunch Program is a government policy that has generated various opinions on social media, ranging from support to criticism regarding its implementation. This study aims to analyze public sentiment before and after the implementation of the Free Lunch Program using the Support Vector Machine (SVM) algorithm and to compare the performance of several kernels, namely Linear, Polynomial, RBF, and Sigmoid. The data were collected from the X (Twitter) platform and classified into three sentiment categories: positive, negative, and neutral. The experimental results show that the RBF kernel achieved the best performance, with an accuracy of 97.04% on pre-implementation data and 96.32% on post-implementation data. Moreover, the sentiment analysis results indicate that the proportion of negative sentiment increased after the program’s implementation, rising from 54.5% to 57.7%. These findings suggest that the RBF kernel is the most effective in classifying public opinions regarding the Free Lunch Program policy with a high level of accuracy, while also illustrating that public perception tends to be more critical after the program was implemented.]]>
Copyrights © 2026
