SENTIMENT ANALYSIS OF COMMENTS ON THE TRAGEDY OF JULIANA MARINS' FALL ON MOUNT RINJANI ON APPLICATION X USING THE MULTINOMIAL NAÏVE BAYES METHOD

Abstract

This study aims to analyze the sentiment of user comments on the X (Twitter) platform regarding the tragic fall of Brazilian hiker Juliana Marins on Mount Rinjani. A total of 1006 comments were collected through a crawling process from June 21, 2025, to July 11, 2025. The research stages include data labeling, text preprocessing (cleaning, case folding, tokenizing, stopword removal, and stemming), N-Gram formation, and feature weighting using TF-IDF. The Multinomial Naïve Bayes algorithm was employed for sentiment classification into three categories: positive, negative, and neutral. Data imbalance was addressed using the Random Oversampling (ROS) technique. Model performance was evaluated using a confusion matrix with accuracy, precision, recall, and F1-Score metrics. The results show that the model achieved an accuracy of 85%, with precision, recall, and F1-Score values indicating effective sentiment classification. Neutral sentiment was found to be the most dominant category among user comments. These findings offer a comprehensive overview of public perception regarding the incident and can serve as a useful reference for decision-making and communication strategies related to similar issues