<![CDATA[Throughout 2024, policies issued by the Ministry of Energy and Mineral Resources (ESDM) concerning fuel triggered diverse public responses on the social media X. This study aims to analyze public sentiment on X toward the policy using 763 tweets collected through web scraping (after removing duplicates from the 913 tweets). Data were labeled using Gemini AI and validated by three annotators, achieving a Fleiss' Kappa of 0.66, indicating substantial agreement. To address class imbalance, back translation was applied as a data augmentation technique, and sentimen classification was performed using Naïve Bayes and Support Vector Machine (SVM) algorithms with N-Gram features. The results show that Naïve Bayes with back translation achieved the best performance, with an F1-score of 85%, compared with 82% for SVM. Topic modeling using Latent Dirichlet Allocation (LDA) identified three main topics with a coherence score of 0,46: the single-gate import policy, fuel scarcity at private gas stations, and the role of the House of Representatives. The dominant negative sentiment indicates an unfavorable public perception of the policy.]]>
Copyrights © 2026
