Sentiment Analysis of Public Opinion on Social Media Regarding the September 17, 2025 Demonstration Using the Support Vector Machine (SVM) Algorithm
Abstract
The development of social media enables people to express opinions about various events, including the September 17, 2025 demonstration involving online motorcycle taxi (ojek online or ojol) drivers. The variety of public responses requires analysis to identify the prevailing sentiment. This study aims to analyze and classify public sentiment toward the September 17, 2025 demonstration using the Support Vector Machine (SVM) algorithm. Data were collected through crawling from X (Twitter) and YouTube using the keywords “demo 17 september 2025” and “#demoojol”, resulting in 1,025 data records. The data underwent preprocessing, including case folding, cleansing, tokenizing, stopword removal, and stemming, followed by labeling into positive, negative, and neutral categories. The data were then weighted using Term Frequency–Inverse Document Frequency (TF-IDF) and divided into 70% training data and 30% testing data for Classification using SVM with a linear kernel. The results showed that negative sentiment was dominant with 643 data (62.7%), followed by positive sentiment with 280 data (27.3%) and neutral sentiment with 102 data (10.0%). The SVM model achieved an accuracy of 70.45%, indicating that it can be used to classify public sentiment toward the September 17, 2025 demonstration.