<![CDATA[Redenomination is a policy of simplifying the nominal value of a currency without changing the exchange rate or purchasing power. The discourse on rupiah redenomination has again become a public concern after the government and Bank Indonesia announced its readiness for implementation, resulting in various public responses on social media, particularly YouTube. This situation has prompted the need for sentiment analysis to accurately identify public opinion trends. This study aims to compare the performance of Naïve Bayes and Support Vector Machine (SVM) in classifying comment sentiment on the channels CNBC Indonesia, Kompas Pontianak, and Kumparan. Data imbalance was handled using the Synthetic Minority Oversampling Technique (SMOTE), while feature extraction used TF-IDF. Evaluation was carried out using accuracy, precision, recall, F1-score, ROC AUC, overfitting testing, and K-Fold Cross Validation. The results show that SMOTE improves the performance of both models. SVM yielded the best results at a 50:50 ratio, with an accuracy of 76.34%, a precision of 76.33%, a recall of 76.34%, an F1-score of 75.46%, and an ROC AUC of 81.44%. Thus, SVM was more optimal than Naïve Bayes in classifying sentiment related to the rupiah redenomination discourse.]]>
Copyrights © 2026
