<![CDATA[Environmental degradation in Sumatra, driven by mining activities and land clearing, has sparked significant public discourse across various digital platforms. However, manual monitoring of massive public sentiment has proven inefficient for policy evaluation purposes. This study aims to classify public sentiment regarding environmental issues based on YouTube comments from a high-level interview involving a Ministry of Forestry advisor on the #IntrigueRK channel. A total of 708 comments were collected via the YouTube API and subjected to comprehensive preprocessing stages, including slang handling and Indonesian language stemming. Classification was performed using the Decision Tree algorithm, selected for its high interpretability in rule-based sentiment mapping. The model achieved an accuracy of 84.4%, with a Precision of 87.5%, a Recall of 89.4%, and an F1-score of 0.843. The research findings reveal a predominance of negative sentiment, reaching 472 comments (66.95%), while positive sentiment accounted for 233 comments (33.05%), primarily focusing on themes of government accountability and the impacts of flash floods. This study demonstrates that the Decision Tree algorithm is a robust instrument for social listening within the environmental sector. These results provide actionable insights for stakeholders to bridge the clinical gap between forestry policy and public perception.]]>
Copyrights © 2026
