YouTube Public Opinion Sentiment Analysis of the Inauguration of Finance Minister Purbaya Using Dictionary Labeling and Decision Trees
Abstract
YouTube social media has become an important platform for Indonesians to express public opinion, including responses to political events such as cabinet minister inaugurations. This study aims to analyze public sentiment on YouTube regarding the inauguration of Finance Minister Purbaya on September 8, 2025, using automatic labeling based on the Indonesian sentiment dictionary (InSet) and testing the performance of the Decision Tree algorithm compared to other classification algorithms. The methods used include collecting 90 YouTube comment data, text preprocessing (cleaning, case folding, tokenizing, stopword removal, and stemming), dictionary-based automatic labeling, TF-IDF feature extraction, and classification using six algorithms: Decision Tree, Support Vector Machine (SVM), K -Nearest Neighbor (KNN), Naïve Bayes, Neural Network, and Deep Learning. The labeling results show the sentiment distribution: 82.2% neutral, 13.3% positive, and 4.4% negative. Model evaluation results show that Decision Tree achieved perfect accuracy (100%) on the test data, outperforming other algorithms that tended to experience bias due to class imbalance. This study concludes that the combination of dictionary-based labeling and the Decision Tree algorithm is effective for sentiment analysis on limited datasets on the YouTube platform, and provides insight that public response to the inauguration of Finance Minister Purbaya was dominated by neutral sentiment.
References
Aribowo, A. S., Basiron, H., Yusof, N. F. A., & Khomsah, S. (2021). Cross-domain sentiment analysis model on indonesian youtube comment. International Journal of Advances in Intelligent Informatics, 7(1), 12–25. https://doi.org/10.26555/ijain.v7i1.554
Armaeni, P. P., Wiguna, I. K. A. G., & Parwita, W. G. S. (2024). Sentiment Analysis of YouTube Comments on the Closure of TikTok Shop Using Naïve Bayes and Decision Tree Method Comparison. Jurnal Galaksi, 1(2), 70–80. https://doi.org/10.70103/galaksi.v1i2.15
Prasetyo, T., Adhyaksa Waskita, A., Taryo, T., Raya Puspitek, J., & Pamulang Kota Tangerang Selatan, K. (2025). Analisis Sentimen Pengguna Seputar Kendaraan Listrik Di Twitter Dengan Penerapan Algoritma Naïve Bayes, KNN, dan Decision Tree untuk Klasifikasi. Jurnal Sistem Komputer Dan Kecerdasan Buatan, 8(2), 108–117. www.dataindonesia.com,
Safra, K. K., & Zuliarso, E. (2025). Analisis Sentimen Terhadap Pelaksanaan Pilkada 2024 Pada Media Sosial Youtube Menggunakan Metode Decision Tree. Jurnal Informatika Teknologi Dan Sains, 7(1), 117–126.
Liu, B. (2020). Sentiment analysis: Mining opinions, sentiments, and emotions (2nd ed.). Cambridge University Press.
Wibowo, A. (2024). Analisis Sentimen Popularitas Capres dan Pilpres pada Media Sosial Twitter: Perbandingan Algoritma SVM, KNN, dan Naïve Bayes. 23(2), 362–374.
Saputra, J., Maryani, L., Wulandari, D., & Eka, W. (2025). Analisis Performa Naive Bayes dan SVM terhadap Sentimen Teks Media Sosial dengan Word2Vec dan SMOTE. JURNAL INSTEK INFORMATIKA SAINS DAN TEKNOLOGI, 10(1), 143–155. https://journal.uin-alauddin.ac.id/index.php/instek
Gede Bagus Resa Pratama Putra, I., Abd Latif, K., Widyawati, L., & Innudin, M. (2025). Analisis Sentimen Berbasis Lexicon Dan Social Network Analysis Terhadap Topik Ruu Tni Pada Media Sosial X. 346–452.
Halimi, A., & Rudyanto Arief, M. (2021). Analisis Sentimen Masyarakat Indonesia Terhadap Pembelajaran Online Dari Di Media Sosial Twitter Menggunakan Lexicon Dan K-Nearest Neighbor. Jurnal Kecerdasan Buatan, Komputasi Dan Teknologi Informasi, 2(1), 18–28. https://ejournal.unuja.ac.id/index.php/core
Mario Conroy Paridy Man, G., Aristo Jansen Sinlae, A., & Ngaga, E. (2025). JIP (Jurnal Informatika Polinema) Analisis Sentimen di Media Sosial X tentang IKN dengan Naïve Bayes. JIP (Jurnal Informatika Polinema), 11, 417–425.
Pramita Widyassari, A., Salsabilla, D., & Ali Amrozi, M. (2025). Analisis sentimen publik di twitter terhadap pelantikan presiden Prabowo menggunakan algoritma Naïve Bayes. Jurnal Ilmiah NERO, 10(1), 2025.
