Komparasi Naïve Bayes dan K-NN Dalam Analisis Sentimen di Twitter Terhadap Kemenangan Paslon 02

Alfira Fitri Nur Azizah (Sistem Informasi, Universitas Merdeka Malang, Indonesia)
Viry Puspaning Ramadhan (Sistem Informasi, Universitas Merdeka Malang, Indonesia)

Journal of Information Technology
Vol 12 No 02 (2024): J-Intech : Journal of Information and Technology

The 2024 Presidential and Vice Presidential Election stands out as a highly awaited political event by the people of Indonesia. The vote counts result, or real count, of the 2024 election have sparked a variety of reactions, both supportive and opposing, especially on social media platforms like Twitter, due to the lead of candidate pair number 02. This study utilizes Twitter as a data source for opinion interpretation. The Naïve Bayes and K-NN were chosen in this study, and their performances are tested and compared. The research results present Naïve Bayes with an accuracy rate of 87.35% +/- 1.81% (micro average: 87.35%), while K-NN algorithm achieved an accuracy rate of 69.68% +/- 3.14% (micro average: 69.68%) using a data partition ratio of 90:10. The analysis results indicate that Naïve Bayes is more effective than K-Nearest Neighbor.