Sentiment Analysis of YouTube Comments on the 2025 DPR RI Demonstration Using Machine Learning

The 2025 DPR RI Demonstration is a national political issue that has triggered a broad response from the Indonesian public, especially through user comments on the YouTube platform. These comments reflect diverse and emotional public opinion, making it relevant to study using a Machine Learning based Sentiment Analysis approach. This study aims to compare the performance of the Support Vector Machine (SVM), Multilayer Perceptron (MLP) based Neural Network (NN), and Random Forest algorithms in classifying YouTube comment sentiments related to the 2025 DPR RI Demonstration issue. Data were obtained through a comment collection process, then processed through text preprocessing and feature weighting stages using the TF-IDF method. The data used in this study were publicly accessible comments, and no personally identifiable information was collected or disclosed to ensure user privacy and ethical data use. Model performance evaluation was conducted using a confusion matrix with accuracy, precision, recall, and F1-score metrics. The results indicate that the SVM algorithm achieved the best performance, with weighted average accuracy, precision, recall, and F1-score values of 96.20%, res