Kajian Opini Publik Terhadap Tayangan Clash of Champions Menggunakan Support Vector Machine

Abstract

This study examines audience responses to the educational program Clash of Champions produced by Ruangguru and inspired by the South Korean format University War. Through its interactive academic competition concept, Clash of Champions has succeeded in attracting public interest, triggering various reactions on social media, and providing a basis for conducting opinion studies to understand audience views. This study uses the Support Vector Machine (SVM) algorithm to classify netizen comments on platform X (Twitter) into positive, negative, and neutral sentiments. The data processing process includes data collection through scraping, text preprocessing, and word weighting using the Term Frequency-Inverse Document Frequency (TF-IDF) algorithm. Model evaluation with accuracy, precision, recall, and F1-score metrics shows various audience opinions, with the majority showing positive sentiments towards the educational innovation of the program. This model achieves an accuracy of 79%, precision of 79%, recall of 95%, and F1-score of 86% for the positive sentiment classification. This indicates a strong appreciation from the public for the educational value of this program. This study is expected to provide insight for educational content developers to create educational and interesting programs.