Sentiment Analysis of Public Comments on Platform X Regarding the Issue of Worker Burnout Using the IndoBERT Base Model

  • Nindita Fatha Dwi Lestari Universitas Jenderal Achmad Yani, Cimahi, Indonesia
  • Fajri Rakhmat Umbara Universitas Jenderal Achmad Yani, Cimahi, Indonesia
  • Melina Universitas Jenderal Achmad Yani, Cimahi, Indonesia
Keywords: Sentiment Analysis, Burnout, IndoBERT, Semi-Supervised Learning, Pseudo-Labeling.

Abstract

Burnout is a mental health problem frequently experienced by workers due to high work pressure, and Platform X users actively share opinions and emotions about it. This study aimed to analyze public sentiment on Platform X regarding worker burnout using IndoBERT Base P2. Data were scraped between June 2023 and June 2025, yielding 2,599 Indonesian-language tweets after filtering. Of these, 300 tweets were labeled manually by the researcher and a psychologist (Cohen’s Kappa 0.8578), and 1,567 more were pseudo-labeled with a confidence threshold of 0.90. Class imbalance was handled with back translation augmentation of the training data (2,769 samples). A grid search of eight experiments identified five epochs and a learning rate of 2e-5 as the best configuration (validation F1-score 0.9797). On the same test set (374 samples), the weighted F1-score increased from 92.88% for the model without semi-supervised learning to 95.50%, together with augmentation and additional training data. Because most of the labeled dataset (1,567 of 1,867 tweets; 83.9%) consisted of pseudo-labels, the accuracy of 95.45% mainly reflects agreement with those labels. Negative sentiment dominated the dataset at 71.34% (a combination of manual labels, pseudo-labels, and model predictions), followed by neutral at 14.93% and positive at 13.74%.

Author Biographies

Fajri Rakhmat Umbara, Universitas Jenderal Achmad Yani, Cimahi, Indonesia

Informatika, Universitas Jenderal Achmad Yani

Melina, Universitas Jenderal Achmad Yani, Cimahi, Indonesia

Informatika, Universitas Jenderal Achmad Yani

References

Abadi, R. R., Abadi, R. R., Ikhwan, M., Haeruddin, M., Fadilah Aswar, N., & Aslam, A. P. (2025). Burnout pada Tenaga Kerja Milenial: Menemukan Harapan Keseimbangan Kehidupan-Kerja. Jurnal ARASTIRMA Universitas Pamulang, 5(1), 302–313.

Afiansyah, M. R., & Adi, A. N. (2024). Pengaruh Beban Kerja dan Stres Kerja Terhadap Burnout Pada Saat COVID-19. Jurnal Kewirausahaan dan Inovasi, 2(2), 543–552. https://doi.org/10.21776/jki.2023.02.2.20

Amini, M. R., Feofanov, V., Pauletto, L., Hadjadj, L., Devijver, E., & Maximov, Y. (2025). Self-training: A survey. Neurocomputing, 616, 128904. https://doi.org/10.1016/j.neucom.2024.128904

Arifudin, A., Khansa, F. B., Mozza, J., Anisa, F., & Khumaerahtuz, Z. (2024). Keseimbangan Kehidupan Kerja Dan Burnout: Implikasinya Terhadap Kepuasan Kerja Karyawan di You N Me Store Cirebon. Jurnal Ekonomi dan Pembangunan, 2(2), 203-217. https://doi.org/10.61132/jepi.v2i2.573

Aryanti, N. N. A., & Suria, O. (2025). Analisis sentimen terhadap pemutusan hubungan kerja di Indonesia: Komparasi IndoBERT dengan SVM, Random Forest, dan Decision Tree dengan optimasi TF-IDF. Rabit: Jurnal Teknologi dan Sistem Informasi Univrab, 10(2), 1158-1176. https://doi.org/10.36341/rabit.v10i2.6364

Azizah, M., Pamungkas, R., & Merdiaty, N. (2024). Kajian Literatur: Burnout & Strategi Penanganannya. Jurnal Psikologi dan Bimbingan Konseling, 9. https://doi.org/10.6734/LIBEROSIS..V2I2.3027

Azmi, A. M., & Al-Ghadir, A. I. (2024). Using Twitter as a digital insight into public stance on societal behavioral dynamics. Journal of King Saud University - Computer and Information Sciences, 36(5). https://doi.org/10.1016/j.jksuci.2024.102078

Chamid, A. A., Widowati, & Kusumaningrum, R. (2024). Labeling Consistency Test of Multi-Label Data for Aspect and Sentiment Classification Using the Cohen Kappa Method. Ingenierie Des Systemes d’Information, 29(1), 161–167. https://doi.org/10.18280/isi.290118

Daffa, M., Fahreza, A., Luthfiarta, A., Rafid, M., Indrawan, M., & Nugraha, A. (2024). Analisis Sentimen: Pengaruh Jam Kerja Terhadap Kesehatan Mental Generasi Z. Journal of Applied Computer Science and Technology (JACOST), 5(1), 2723–1453. https://doi.org/10.52158/jacost.715

Devlin, J., Chang, M. W., Lee, K., & Toutanova, K. (2019, June). Bert: Pre-training of deep bidirectional transformers for language understanding. In Proceedings of the 2019 conference of the North American chapter of the association for computational linguistics: human language technologies, volume 1 (long and short papers) (pp. 4171-4186). https://github.com/tensorflow/tensor2tensor

Fadhillah, N. (2023). Cyberspace Sebagai Ruang Publik Baru: Analisis Naratif Opini Publik Pada Isu Larangan Thrifting di Media Sosial Twitter. Jurnal Ilmu Komunikasi Pattimura, 02(01).

Grandini, M., Bagli, E., & Visani, G. (2020). Metrics for Multi-Class Classification: an Overview. http://arxiv.org/abs/2008.05756

Homepage, J., Musfiroh, D., Khaira, U., Eko, P., Utomo, P., Suratno, T., Studi, P., Informasi, S., Sains, F., & Teknologi, D. (2021). Sentiment Analysis of Online Lectures in Indonesia from Twitter Dataset Using InSet Lexicon Analisis Sentimen terhadap Perkuliahan Daring di Indonesia dari Twitter Dataset Menggunakan InSet Lexicon. MALCOM: Indonesian Journal of Machine Learning and Computer Science 1, 24–33.

Jeevallucas, P., Surya Gautama, J., Wibowo, A., Siang, J. J., Informasi, S., Informasi, T., Kristen, U., & Wacana, D. (2025). Analisis Sentimen Pengguna Terhadap Akun X/Twitter Resmi “Dana” dengan Algoritma Indobert Sentiment Analysis of Users Toward the Official “Dana” Account on X/Twitter Using the Indobert Algorithm. Journal of Information Technology and Computer Science (INTECOMS), 8(2).

Kamil, A. R., Lianda, I. C., Silaban, N. Y., & Wulansari, N. (2023). Identifikasi Burnout Syndrome Perawat Di Lingkungan Aliansi RS Islam Jakarta. Jurnal Ilmiah Keperawatan IMELDA, 9(2), 115-121.

Koto, F., Rahimi, A., Lau, J. H., & Baldwin, T. (2020, December). IndoLEM and IndoBERT: A benchmark dataset and pre-trained language model for Indonesian NLP. In Proceedings of the 28th International Conference on Computational Linguistics (pp. 757-770). http://arxiv.org/abs/2011.00677

Pratama, A. Y., Sanjaya, G. A., Lubis, N. K., Aditya, M. R., & Yennimar, Y. (2025). Analisis Sentimen Publik Terkait Danantara Menggunakan Algoritma IndoBERT pada Platform Media Sosial. Metik Jurnal, 9(1), 92-100. https://doi.org/10.47002/metik.v9i1.1055

Ramadhani, R., Juandi, D., & Nurlaelah, E. (2021). A Meta-Analysis on The Effect of Inquiry Learning Model on Students’ Mathematical Problem-Solving Skills. Indonesian Journal of Science and Mathematics Education, 4(3), 302–312. https://doi.org/10.24042/ijsme.v4i3.9730

Salma, T. D., Kurniawan, M. F., Darmawan, R., & Basri, A. (2025). Analisis Sentimen Berbasis Transformer: Persepsi Publik terhadap Nusantara pada Perayaan Kemerdekaan Indonesia yang Pertama. vol, 9, 757-764.

Simanjuntak, A., Lumbantoruan, R., Sianipar, K., Gultom, R., Simaremare, M., Situmeang, S., & Panggabean, E. (2024). Research and analysis of indobert hyperparameter tuning in fake news detection. Jurnal Nasional Teknik Elektro dan Teknologi Informasi, 13(1), 60-67. https://doi.org/10.22146/jnteti.v13i1.8532

Society for Human Resource Management (SHRM), “SHRM 2025 Insights Workplace Mental Health,” 2025. (2025).

van Engelen, J. E., & Hoos, H. H. (2020). A survey on semi-supervised learning. Machine Learning, 109(2), 373–440. https://doi.org/10.1007/s10994-019-05855-6

Wildan, A. H., & Nastiti, V. R. S. (2024). Perbandingan kinerja pre-trained IndoBERT-base dan IndoBERT-lite pada klasifikasi sentimen ulasan TikTok Tokopedia Seller Center dengan model IndoBERT. Jsii (Jurnal Sist. Informasi), 11(2), 13-20. https://doi.org/10.30656/jsii.v11i2.9168

Published
2026-11-15
How to Cite
Lestari, N. F. D., Umbara, F. R., & Melina. (2026). Sentiment Analysis of Public Comments on Platform X Regarding the Issue of Worker Burnout Using the IndoBERT Base Model. International Journal of Science and Society, 8(4), 1000-1016. https://doi.org/10.54783/ijsoc.v8i4.1853