Comparative Sentiment Analysis of Indonesian Social Media Opinions on Fuel Subsidy Policy Using IndoBERT and NusaBERT
DOI:
https://doi.org/10.30871/jaic.v10i3.12964Keywords:
BBM Policy, IndoBERT, NusaBERT, Public Sentiment, TransformerAbstract
Rising geopolitical tensions in the Middle East have triggered public concerns in Indonesia regarding fuel subsidy policies and fuel availability. This study aims to compare the performance of IndoBERT and NusaBERT in classifying Indonesian public sentiment on social media related to fuel subsidy policies. Data were collected from X (Twitter) and Instagram comments between October and November 2025 using keywords such as “BBM”, “Pertalite”, “fuel subsidy”, and “Middle East conflict”. After filtering duplicate, spam, and irrelevant content, a total of 1,500 opinion texts were manually annotated into positive, neutral, and negative sentiment classes and divided using an 80:20 train-test split configuration. The preprocessing stage included case folding, text cleansing, slang word normalization, emoji removal, duplicate filtering, and tokenization. Both Transformer models were fine-tuned using the AdamW optimizer with a learning rate of 2e-5, batch size of 16, and 3 training epochs. Model performance was evaluated using accuracy, precision, recall, and F1-score metrics. Experimental results show that NusaBERT achieved better performance than IndoBERT, obtaining an accuracy of 96.3% and weighted F1-score of 96.3%, while IndoBERT achieved an accuracy of 92.5%. Additional evaluation through confusion matrix and error analysis indicates that both models still face challenges in handling sarcasm, ambiguous expressions, and mixed-context sentences commonly found in informal Indonesian social media text. The findings suggest that NusaBERT is more effective for Indonesian social media sentiment classification due to its stronger adaptation to informal language patterns.
Downloads
References
[1] C. D. Aprida, “Menyulam Ketegangan Timur Tengah Dampak Ekonomi Global dari Konflik Iran-Israel,” vol. 1, no. 1, pp. 25–36, 2025.
[2] D. Soesanto Edy, Utami Dewi Puspita, “Dampak Fluktuasi Harga Minyak Dunia Terhadap Ekonomi di Indonesia,” Ris. Ilm., vol. 2, no. 1, pp. 231–242, 2025, doi: https://doi.org/10.62335 Jurnal.
[3] E. M. Dilasari, “Pengaruh Harga Minyak Goreng dan Panic Buying Terhadap Keputusan Pembelian dalam Perspektif Bisnis Syariah,” 2023.
[4] J. Jtik, J. Teknologi, and M. Samantri, “Perbandingan Algoritma Support Vector Machine dan Random Forest untuk Analisis Sentimen Terhadap Kebijakan Pemerintah Indonesia Terkait Kenaikan Harga BBM Tahun 2022,” vol. 8, no. 1, pp. 1–9, 2024.
[5] N. Wulandari, Y. Cahyana, and H. H. Handayani, “Sentiment Analysis on the Relocation of the National Capital ( IKN ) on Social Media X Using Naive Bayes and K-Nearest Neighbor ( KNN ) Methods,” vol. 9, no. 3, pp. 724–731, 2025.
[6] B. Wilie et al., “IndoNLU : Benchmark and Resources for Evaluating Indonesian Natural Language Understanding,” pp. 843–857, 2020.
[7] S. Cahyawijaya et al., “NusaCrowd : Open Source Initiative for Indonesian NLP Resources,” 2022.
[8] Liu, M. Ott, N. Goyal, et al., “RoBERTa: A Robustly Optimized BERT Pretraining Approach,” arXiv preprint arXiv:1907.11692, 2019.
[9] M. C. Kenton, L. Kristina, and J. Devlin, “BERT : Pre-training of Deep Bidirectional Transformers for Language Understanding,” pp. 4171–4186, 2019.
[10] N. Nur, A. Aryanti, and O. Suria, “Analisis Sentimen Terhadap Pemutusan Hubungan Kerja di Indonesia : Komparasi IndoBERT dengan SVM,Random Forest, dan Decision Treedengan Optimasi TF-IDF,” Teknol. dan Sist. Inf. Univrab, vol. 10, no. 2, pp. 1158–1176, 2025.
[11] P. Subarkah, A. N. Ikhsan, E. Anggraeni, and A. P. Sabaniyah, “Sentiment Perspective of Government ’ s Free Nutritious Meal Policy on Social Media X using Indo-BERT and Bi-LTSM,” vol. 7, no. 2, pp. 201–210, 2025.
[12] M. Y. Dhinora and E. Mailoa, “Analisa Tweet Mahasiswa untuk Deteksi Gejala Depresi dengan Penerapan Natural Language Processing,” vol. 6, no. 2, pp. 1193–1211, 2025.
[13] W. Syahrina, “Strategi Komunikasi Public Relations dalam Mengelola Isu Lingkungan pada Perusahaan Energi Syahrina Wirdani,” Adm. Publik Univ. Medan Area, pp. 1–8.
[14] A. W. V Hutabarat, N. Luh, S. Saraswati, and K. Suryadi, “Analisis Sentimen Data Ulasan Pengguna MyPertamina di Twitter dengan Metode Machine Learning dan Deep Learning,” vol. 13, no. 1, pp. 145–154, 2024.
[15] L. Afuan, N. Hidayat, H. Hamdani, and H. Ismanto, “Optimizing BERT Models with Fine-Tuning for Indonesian Twitter Sentiment Analysis,” vol. 2, pp. 248–267, 1959, doi: 10.58346/JOWUA.2025.I2.016.
[16] E. Bagli and G. Visani, “Metrics for multi-class classification: An overview,” pp. 1–17, 2020.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Dini Ambarwati, Pungkas Subarkah, Septi Nurhayati
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:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License (Attribution-ShareAlike 4.0 International (CC BY-SA 4.0) ) that allows others to share the work with an acknowledgement 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 acknowledgement 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 (See The Effect of Open Access).
