Implementasi Analisis Sentimen Komentar YouTube Berbasis IndoBERT dengan Evaluasi Confidence Score
DOI:
https://doi.org/10.29407/vvdxca28Keywords:
Analisis Sentimen, Confidence Score, IndoBERT, Komentar Youtube, TranformerAbstract
Analisis sentimen terhadap opini publik di media sosial menjadi penting untuk memahami persepsi masyarakat terhadap suatu topik, khususnya dalam konteks olahraga nasional seperti Timnas Indonesia. Penelitian ini mengimplementasikan sistem analisis sentimen otomatis pada komentar YouTube menggunakan pre-trained model IndoBERT (Indonesian Bidirectional Encoder Representations from Transformers) untuk mengklasifikasikan sentimen ke dalam tiga kategori: positif, negatif, dan netral. Data dikumpulkan melalui YouTube Data API v3 dari empat video terkait Timnas Indonesia, menghasilkan 952 komentar unik setelah melalui tahap preprocessing yang mencakup case folding, text cleaning (penghapusan URL, mentions, hashtags), dan deduplication. Pelabelan sentimen dilakukan secara otomatis menggunakan model w11wo/indonesian-roberta-base-sentiment-classifier dengan batch processing. Evaluasi sistem menggunakan confidence score sebagai metrik internal menunjukkan performa yang sangat baik dengan rata-rata confidence 89.68%, konsisten across semua kategori sentimen (positif: 89.47%, negatif: 89.76%, netral: 89.75%). Hasil klasifikasi menunjukkan dominasi sentimen negatif (43.59%), diikuti netral (29.52%) dan positif (26.89%), mengindikasikan opini publik yang cenderung kritis terhadap Timnas Indonesia dalam periode observasi. High confidence score distribution (71.4% prediksi ≥90%) menunjukkan bahwa model mampu mengidentifikasi sentiment signals dengan akurat pada mayoritas kasus, menjadikan pendekatan ini feasible untuk real-time sentiment monitoring pada social media discourse.
Downloads
References
[1] A. Hussain, A Practical Guide to Sentiment Analysis, vol. 5. Springer, 2017. doi: 10.1007/978-3-319-55394-8.
[2] J. Devlin, M.-W. Chang, K. Lee, K. T. Google, and A. I. Language, “BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding,” Proceedings of the 2019 Conference of the North, pp. 4171–4186, 2019, doi: 10.18653/V1/N19-1423.
[3] B. Wilie et al., “IndoNLU: Benchmark and Resources for Evaluating Indonesian Natural Language Understanding,” pp. 843–857, Nov. 2020, doi: 10.18653/V1/2020.AACL-MAIN.85.
[4] C. Guo, G. Pleiss, Y. Sun, and K. Q. Weinberger, “On Calibration of Modern Neural Networks,” 34th International Conference on Machine Learning, ICML 2017, vol. 3, pp. 2130–2143, Jun. 2017, Accessed: Dec. 31, 2025. [Online]. Available: https://arxiv.org/pdf/1706.04599
[5] O. El Azzouzy, T. Chanyour, and S. J. Andaloussi, “Transformer-based models for sentiment analysis of YouTube video comments,” Sci Afr, vol. 29, p. e02836, Sep. 2025, doi: 10.1016/J.SCIAF.2025.E02836.
[6] E. Haddi, X. Liu, and Y. Shi, “The Role of Text Pre-processing in Sentiment Analysis,” Procedia Comput Sci, vol. 17, pp. 26–32, Jan. 2013, doi: 10.1016/J.PROCS.2013.05.005.
[7] O. El Azzouzy, T. Chanyour, and S. J. Andaloussi, “Transformer-based models for sentiment analysis of YouTube video comments,” Sci Afr, vol. 29, p. e02836, Sep. 2025, doi: 10.1016/J.SCIAF.2025.E02836.
[8] “Pattern Recognition and Machine Learning | Springer Nature Link (formerly SpringerLink).” Accessed: Dec. 31, 2025. [Online]. Available: https://link.springer.com/book/9780387310732
[9] “Deep Learning.” Accessed: Dec. 31, 2025. [Online]. Available: https://www.deeplearningbook.org/
[10] C. Guo, G. Pleiss, Y. Sun, and K. Q. Weinberger, “On Calibration of Modern Neural Networks,” 34th International Conference on Machine Learning, ICML 2017, vol. 3, pp. 2130–2143, Jun. 2017, Accessed: Dec. 31, 2025. [Online]. Available: https://arxiv.org/pdf/1706.04599
[11] J. D. Kelleher, B. M. Namee, and A. D’Arcy, “Machine learning for predictive data analytics,” MIT Press, p. 43, 2001, Accessed: Dec. 31, 2025. [Online]. Available: https://mitpress.mit.edu/9780262029445/fundamentals-of-machine-learning-for-predictive-data-analytics/
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Aji Pangestu, Frisqy Maulana
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Copyright on any article is retained by the author(s).
- The author grants the journal, right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgment of the work’s authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal’s published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work.
- The article and any associated published material is distributed under the Creative Commons Attribution-ShareAlike 4.0 International License
