ANALISIS SENTIMEN KOMENTAR NETIZEN PADA MEDIA SOSIAL INSTAGRAM TIMNAS INDONESIA MENGGUNAKAN METODE NAÏVE BAYES DAN SUPPORT VECTOR MACHINE

Abstract

Social media has become an integral part of modern society, providing a platform for public expression. In Indonesia, soccer is a popular sport, and the Indonesian national team is a major topic of discussion on social media, especially Instagram. This study aims to analyze the sentiment of netizen comments on the official Instagram account of the Indonesian national team (@timnasindonesia) using the Naïve Bayes and Support Vector Machine (SVM) methods. The data used consisted of 2,920 comments from 50 posts between September 28, 2023, and May 10, 2024, focusing on the U-23 and Senior National Team matches. Naïve Bayes and SVM were chosen because both are effective methods for sentiment classification. Naïve Bayes uses a probabilistic approach, while SVM seeks the best hyperplane to separate data classes. The results of the study show that both methods can be used to analyze sentiment with a good level of accuracy. Testing results on each training and testing data with different presentations will give different accuracy results. The test results from the Naive Bayes method obtained the highest accuracy value of 69% with a ratio of 70:30. Meanwhile, the Support Vector Machine (SVM) method obtained the highest accuracy of 74% at the same ratio. It can be concluded that the Support Vector Machine method obtained a higher accuracy value than the Naive Bayes method, making it more optimal for use in classifying the sentiment of netizen comments in this study.