<![CDATA[This study aims to compare the performance of the Random Forest and Support Vector Machine (SVM) algorithms in conducting sentiment analysis on YouTube comments related to education reform in Indonesia. The dataset used in this study consisted of 981 comments collected from the YouTube platform, and randomly labeled with two categories: "positive" and "negative." The labeling process was carried out using Microsoft Excel, while data processing was carried out using RapidMiner software. Model evaluation was carried out using the cross-validation method to obtain more objective results and avoid overfitting. The results showed that the Random Forest algorithm obtained an accuracy of 99.87% ± 0.40% with a micro average of 99.87%, while the SVM algorithm produced an accuracy of 90.58% ± 3.78% with a micro average of 90.57%. Based on these results, it can be concluded that Random Forest has superior performance in classifying comment sentiment compared to SVM. This is due to Random Forest's ability to combine several decision trees to produce more stable and accurate predictions. The findings of this study can be a reference for other researchers in selecting the right algorithm for sentiment analysis on text data, especially in the context of education and public opinion.]]>
Copyrights © 2026
