<![CDATA[Indonesian Ministry of Education, Culture, Research, and Technology introduced the Merdeka Curriculum to overcome post-pandemics COVID-19 challenges with a transformative education policy. However, the effectiveness of education policy requires the contribution of public opinion. Exploring sentiment public opinions, and structural dimensions of interaction or network behavior are needed. This paper proposed sentiment and social network analysis of X social media on the implementation of the Merdeka Curriculum in Indonesia. The methodology consists of preprocessing, vectorizer with TF-IDF, sentiment and social network analysis. Sentiment classification was performed using four machine learning models, Support Vector Machine (SVM), Random Forest, XGBoost, and LightGBM. Based on the experimental results, Random Forest achieved the best performance with 70% accuracy. The analysis revealed that public sentiment was dominated by neutral and negative responses, indicating persistent criticism and limited support throughout the curriculum’s rollout. Social network analysis identified central accounts in the discourse, including @nadiemmakarim and @Kemdikbud_RI, while other accounts, for example @adekumala, served as key bridges within the network despite receiving fewer mentions. This paper integrates a data-driven approach to understanding public opinion and shows influence dynamics on social media providing valuable insights for policy communication and refinement in the digital era.]]>
Copyrights © 2025
