Analisis Sentimen Pengguna Instagram Terkait Pemiliha Presiden 2024 Dengan Metode Naive Bayes Classifier (Nbc) Dan Support Vector Machine (Svm) (Studi Kasus : Penggunaan Opini Publik Di Media Sosial)

Abstract

n the context of the 2024 presidential election, social media has become the primary platform where public opinions are expressed. However, many news articles highlight the high level of negative opinions that characterize this election. To address this issue, sentiment analysis becomes an effective solution utilizing classification methods such as Naive Bayes Classifier (NBC) and Support Vector Machine (SVM). This study explores sentiment analysis of Instagram users related to the 2024 presidential election. Data consisting of 5.717 Instagram comments were collected to evaluate public opinions. The analysis results show that SVM with RBF kernel achieves an 97.89% accuracy, while NBC achieves an 82.39% accuracy on testing data. The analysis also indicates an increase in the number of positive comments after sentiment analysis. This research provides a deeper understanding of public opinion on social media regarding the 2024 presidential election, and compares the performance of two commonly used classification methods. The findings can serve as a guide for political stakeholders to better understand and respond to public opinion more effectively through social media.