The Public Sentiment Analysis on the 2024 Presidential Election Using Naive Bayes Classifier (NBC) and Support Vector Machine (SVM) On Social Media Data

Asro Asro (Universitas Raharja, Tangerang)
Nur Azizah (Universitas Raharja, Tangerang)
Sudaryono Sudaryono (Universitas Raharja, Tangerang)

Prosiding Amal Insani Foundation
Vol. 1 (2024): PROSIDING INTERNASIONAL

This study aims to evaluate the effectiveness of the Naive Bayes Classifier (NBC) and Support Vector Machine (SVM) in analyzing public sentiment from YouTube comments related to the 2024 Indonesian Presidential Election. A total of 1,800 comments, collected from November 2023 to March 2024, were analyzed to test these models. The results show that SVM, with the highest accuracy of 76.33% and precision and F1-Score of 75.29% and 72.67% on the 10% test data, outperformed NBC, which recorded a highest accuracy of 72.19% under similar conditions. These findings highlight the importance of using more sophisticated methods in sentiment analysis to understand the complex and diverse dynamics of public opinion. This study provides valuable insights for stakeholders in developing effective communication strategies and offers a foundation for advancing sentiment analysis methodologies in political contexts.