<![CDATA[The development of social media, particularly Twitter, has positioned it as a primary platform for expressing public opinion on political issues in Indonesia. One phenomenon that has attracted significant attention is the viral hashtag #BubarkanDPR, which reflects increasing public criticism of the performance of the legislative institution. Several previous studies have shown that the Naïve Bayes machine learning method performs well in sentiment classification tasks. A review of five relevant journals reveals varying accuracy levels: (1) tweet-based sentiment analysis on corruption achieved an accuracy of 82–87%, (2) sentiment analysis of anti-corruption campaigns reached 84%, (3) research on public sentiment toward the Corruption Eradication Commission (KPK) showed a Naïve Bayes accuracy of 82%, (4) a study on the revision of the KPK Law reported an accuracy of 78%, and (5) a comparative study of methods on corruption and tax issues recorded an accuracy of 80% for Naïve Bayes. These findings confirm that Naïve Bayes is consistently applied to political and sensitive topics with stable performance. This study examines public sentiment toward the hashtag #BubarkanDPR by applying the Naïve Bayes method. Data were collected through crawling Twitter comments and processed through several stages, including cleaning, case folding, tokenization, stopword removal, and stemming. The model was evaluated using a confusion matrix. The results show that the model achieved an accuracy rate of 77%, which is consistent with the accuracy range reported in several previous studies. Thus, Naïve Bayes is proven to be sufficiently effective in analyzing sentiment on dynamic and controversial political issues. This study provides insights into public perception and can serve as a reference for further research on social media–based public opinion analysis.]]>
Copyrights © 2026
