Cross-Lingual Sentiment Analysis for Indonesian Monetary Policy

Akbar Ramadhan; Umar Zaky

doi:10.58526/jsret.v4i4.943

Authors

Akbar Ramadhan University of Technology Yogyakarta
Umar Zaky Universitas Teknologi Yogyakarta

DOI:

https://doi.org/10.58526/jsret.v4i4.943

Keywords:

Cross-Lingual Sentiment Analysis, Auto-Labeling, Weak Supervision, RoBERTa, IndoBERT, Back-Translation, Monetary Policy, Indonesia

Abstract

This research develops a cross-lingual sentiment analysis system (RoBERTa-IndoBERT) to monitor public opinion on Bank Indonesia’s 2025 monetary policy from X (Twitter), addressing the scarcity of Indonesian labels and noisy social media text. We introduce a "translate-then-classify" pipeline: Indonesian posts are translated into English, auto-labeled by a mature English RoBERTa model, and these labels are used to fine-tune IndoBERT on the original texts. We compare this cross-lingual (CL) approach, with and without back-translation (BT) augmentation, against a baseline Indo-only model. Performance measured by Accuracy and Macro-F1 indicates the CL pipeline is substantially better than the baseline. The complete model (IndoBERT + CL + BT) yields a Macro-F1 of 98.1%, a 2.8 percentage point (pp) improvement over the baseline (95.3%). Qualitative error analysis corroborates the CL model is more stable, less prone to extreme polarity flips, and better at detecting implicit sentiment. This research demonstrates that a CL auto-labeling pipeline is an efficient and resilient solution for Indonesian sentiment analysis in low-resource scenarios.

Downloads

Download data is not yet available.

Author Biographies

Akbar Ramadhan, University of Technology Yogyakarta

Akbar Ramadhan is an Undergraduate Student (or Student) in the Informatics Engineering Department at Universitas Teknologi Yogyakarta. Research focus is on Natural Language Processing (NLP).

Umar Zaky, Universitas Teknologi Yogyakarta

Umar Zaky is a Lecturer in the Information Systems Department at Universitas Teknologi Yogyakarta.

References

Czudaj, R. L., & Nguyen, B. N. (2025). ECB’s central bank communication and monetary policy transmission: predictability from text-based sentiment indicators? Macroeconomic Dynamics, 29. https://doi.org/10.1017/S1365100525000239

Feng, S. Y., Gangal, V., Wei, J., Chandar, S., Vosoughi, S., Mitamura, T., & Hovy, E. (2021). A Survey of Data Augmentation Approaches for NLP. Computer Science, 968–988. https://arxiv.org/abs/2105.03075

Gaurav, A., Gupta, B. B., Sharma, S., Bansal, R., & Chui, K. T. (2024). XLM-RoBERTa Based Sentiment Analysis of Tweets on Metaverse and 6G. Procedia Computer Science, 238, 902–907. https://doi.org/10.1016/j.procs.2024.06.110

Gorodnichenko, Y., Pham, T., & Talavera, O. (2025). Central bank communication on social media: What, to whom, and how? Journal of Econometrics, 249. https://doi.org/10.1016/j.jeconom.2024.105869

Kemp, S. (2024, February 21). Digital 2024: Indonesia. DATAREPORTAL. https://datareportal.com/reports/digital-2025-sub-section-top-social-platforms?

Koto, F., Rahimi, A., Lau, J. H., & Baldwin, T. (2020). IndoLEM and IndoBERT: A Benchmark Dataset and Pre-trained Language Model for Indonesian NLP. Online. https://huggingface.co/

Li, B., Hou, Y., & Che, W. (2022). Data augmentation approaches in natural language processing: A survey. AI Open, 3, 71–90. https://doi.org/10.1016/j.aiopen.2022.03.001

Liu, Y., Ott, M., Goyal, N., Du, J., Joshi, M., Chen, D., Levy, O., Lewis, M., Zettlemoyer, L., & Stoyanov, V. (2019). RoBERTa: A Robustly Optimized BERT Pretraining Approach. http://arxiv.org/abs/1907.11692

Manoppo, M. R., Kolang, I. C., Nur Fiat, D. N., Mawara, R. M. C., Sumarno, A. D. P., Yusupa, A., & Tarigan, V. (2025). ANALISIS SENTIMEN PUBLIK DI MEDIA SOSIAL TERHADAP KENAIKAN PPN 12% DI INDONESIA MENGGUNAKAN INDOBERT. Jurnal Kecerdasan Buatan Dan Teknologi Informasi, 4(2), 152–163. https://doi.org/10.69916/jkbti.v4i2.322

Pellicer, L. F. A. O., Ferreira, T. M., & Costa, A. H. R. (2023). Data augmentation techniques in natural language processing. Applied Soft Computing, 132. https://doi.org/10.1016/j.asoc.2022.109803

Picault, M., Pinter, J., & Renault, T. (2022). Media sentiment on monetary policy: determinants and relevance for inflation expectations.

Pramono, B. (2025, June 18). BI-Rate Held at 5.50% Maintaining Stability, Strengthening Economic Growth. Bank Indonesia. https://www.bi.go.id/en/publikasi/ruang-media/news-release/Pages/sp_2713325.aspx?utm_source=chatgpt.com

Přibáň, P., Šmíd, J., Steinberger, J., & Mištera, A. (2024). A comparative study of cross-lingual sentiment analysis. Expert Systems with Applications, 247(C). https://www.sciencedirect.com/science/article/abs/pii/S095741742400112X

Reuters. (2024, June 14). Indonesia c.bank intervenes to defend faltering rupiah. Reuters. https://www.reuters.com/business/finance/indonesia-cbank-says-policy-aims-ensure-inflation-controlled-rupiah-stable-2024-06-14/?utm_source=chatgpt.com

Semary, N. A., Ahmed, W., Amin, K., Pławiak, P., & Hammad, M. (2023). Improving sentiment classification using a RoBERTa-based hybrid model. Frontiers in Human Neuroscience, 17. https://doi.org/10.3389/fnhum.2023.1292010

Setiawan, D., Utari Iswavigra, D., & Anggiratih, E. (2025). Implementation of IndoBERT for Sentiment Analysis of the Constitutional Court’s Decision Regarding the Minimum Age of Vice Presidential Candidates. Scientific Journal of Informatics, 12(3). https://doi.org/10.15294/sji.v12i3.26360

Xie, Q., Dai, Z., Hovy, E., Luong, M.-T., & Le, Q. V. (2020). Unsupervised Data Augmentation for Consistency Training. https://github.com/google-research/uda.

Xu, Y., Cao, H., Du, W., & Wang, W. (2022). A Survey of Cross-lingual Sentiment Analysis: Methodologies, Models and Evaluations. In Data Science and Engineering (Vol. 7, Issue 3, pp. 279–299). Springer. https://doi.org/10.1007/s41019-022-00187-3