<![CDATA[In the digital era, public opinion spreads massively and instantly through various social media platforms. One issue that has sparked widespread attention and debate in the digital space is regarding the authenticity of President Joko Widodo's diploma. This issue has provoked various reactions, both support, criticism, and neutral attitudes from netizens, especially through the comments column on YouTube. This study analyzes public sentiment towards the issue using a machine learning approach with the Naïve Bayes and K-Nearest Neighbors (KNN) algorithms, as well as the SMOTE data balancing technique. A total of 1,000 comments were analyzed and classified into three sentiment categories, namely positive, negative, and neutral. Four test scenarios were carried out, namely: KNN, KNN with SMOTE, Naïve Bayes, and Naïve Bayes with SMOTE with a performance comparison tested to see the effectiveness of each in classifying digital opinion. The test results showed that the combination of Naïve Bayes and SMOTE provided the best performance with accuracy, precision, recall, and F1-score of 73%. In contrast, the worst performing model is KNN with SMOTE, which only achieves 27% accuracy, 53% precision, 34% recall, and 15% F1-score. This study emphasizes the importance of algorithm selection and data handling strategies in digital opinion classification, and can be the basis for developing a reliable sentiment analysis system in the future.]]>
Copyrights © 2025
