<![CDATA[The rapid development of social media has created new spaces for the public to express opinions regarding various public policies, including the development of Indonesia’s new capital city, Nusantara (IKN). TikTok, as one of the platforms with high user interaction, provides a valuable data source for analyzing public perceptions of this national development project. This study aims to analyze sentiment in TikTok comments related to the progress of IKN development and to compare the performance of the Naïve Bayes and Support Vector Machine (SVM) classification algorithms. The research employs a quantitative approach using a data mining framework based on the SEMMA methodology, which includes the stages of Sample, Explore, Modify, Model, and Assess. The dataset was collected through web scraping using Apify, resulting in 2,000 comments, of which 1,850 valid comments remained after the selection process. Text preprocessing was performed through cleaning, case folding, tokenizing, stopword removal, and filtering, followed by feature extraction using the TF-IDF method. The dataset was divided into training and testing sets using an 80:20 ratio. Model evaluation was conducted using a confusion matrix with accuracy, precision, recall, and F1-score metrics. The results show that the SVM algorithm outperformed Naïve Bayes with an accuracy of 91.25%, precision of 90.70%, recall of 92.86%, and F1-score of 91.77%, while Naïve Bayes achieved an accuracy of 84.25%, precision of 83.87%, recall of 86.67%, and F1-score of 85.24%. The sentiment distribution indicates that positive sentiment toward the development of IKN slightly dominates negative sentiment. These findings suggest that SVM is more effective for classifying sentiment in informal social media text such as TikTok comments.]]>
Copyrights © 2026
