Explainable Aspect-Based Sentiment Analysis with Contrast-Aware IndoBERT for Indonesian Public Service Reviews
-
Muhammad Shihab Fathurrahman Jondien
*
Amikom Purwokerto University, Purwokerto,
Indonesia
- Taqwa Hariguna Amikom Purwokerto University, Purwokerto, Indonesia
- Dhanar Intan Surya Saputra Amikom Purwokerto University, Purwokerto, Indonesia (*) Corresponding Author
Abstract
This study presents an Explainable IndoBERT with Contrast-Aware Attention framework for Aspect-Based Sentiment Analysis (ABSA) on Indonesian public service reviews. The proposed model integrates automated aspect labeling using KeyBERT with a contrast-aware mechanism to handle mixed or opposing sentiments within a single sentence. By leveraging IndoBERT as the base transformer, the system captures context-sensitive sentiment cues while maintaining interpretability through attention-based rationale extraction. Experimental results on the SMSA dataset demonstrate an accuracy of 83.4%, with strong precision in positive and negative sentiment detection. The contrast-aware module improves clause-level understanding, while the attention-based explainability module provides transparent, token-level rationales that align with human judgments at an average rate of 87.7%. Although a modest performance decline occurs compared to non-explainable baselines, the proposed model offers significant gains in semantic transparency, making it suitable for evidence-based policy evaluation and citizen feedback monitoring. This research contributes a practical, interpretable, and linguistically grounded solution for explainable sentiment analysis in low-resource languages, advancing the application of responsible AI in public service analytics.
Downloads
References
A. Ruelens, “Analyzing user-generated content using natural language processing: A case study of public satisfaction with healthcare systems,” J. Comput. Social Sci., vol. 5, no. Jun., pp. 731–749, 2021, doi: 10.1007/s42001-021-00148-2.
D. I. Putri, A. N. Alfian, M. Y. Putra, and P. D. Mulyo, “IndoBERT model analysis: Twitter sentiments on Indonesia’s 2024 presidential election,” J. Appl. Informatics Comput. (JAIC), vol. 8, no. 1, pp. 7–12, Jan. 2024, doi: 10.30871/jaic.v8i1.7440.
D. Jayakody, K. Isuranda, A. Malkith, N. de Silva, S. R. Ponnamperuma, G. Sandamali, and K. L. Sudheera, “Aspect-based sentiment analysis techniques: A comparative study,” Moratuwa Eng. Res. Conf., vol. 2024, no. Jul., pp. 205–210, 2024, doi: 10.1109/mercon63886.2024.10688631.
A. Maroof, S. Wasi, S. I. Jami, and M. S. Siddiqui, “Aspect-based sentiment analysis for service industry,” IEEE Access, vol. 12, no. Jun., pp. 109702–109713, 2024, doi: 10.1109/access.2024.3440357.
G. Brauwers and F. Frasincar, “A survey on aspect-based sentiment classification,” ACM Comput. Surveys, vol. 55, no. May, pp. 1–37, 2021, doi: 10.1145/3503044.
I. Surjandari, R. A. Wayasti, Z. Zulkarnain, E. Laoh, A. M. M. Rus, and I. Prawiradinata, “Mining public opinion on ride-hailing service providers using aspect-based sentiment analysis,” Int. J. Technol., vol. 10, no. 4, pp. 697–706, 2019, doi: 10.14716/ijtech.v10i4.2860.
A. Jazuli, “Optimizing aspect-based sentiment analysis using BERT for Indonesian student reviews,” Appl. Sci., vol. 15, no. 1, pp. 1–10, Jan. 2024, doi: 10.3390/app15010172.
F. R. Andhika, “Analisis sentimen menggunakan IndoBERT dan GloVe untuk ulasan aplikasi Zoom,” METIK J. Informatics Technol., vol. 9, no. 2, pp. 25–35, May 2025, doi: 10.33096/metik.v9i2.1098.
M. Fuadi, A. Wibawa, and S. Sumpeno, “idT5: Indonesian version of multilingual T5 transformer,” arXiv preprint, vol. 2023, no. Feb., pp. 1–10, 2023, doi: 10.48550/arxiv.2302.00856.
U. Khairani, V. Mutiawani, and H. Ahmadian, “Pengaruh tahapan preprocessing terhadap model IndoBERT dan IndoBERTweet pada deteksi emosi komentar Instagram,” J. Teknol. Inf. Ilmu Komput. (JTIik), vol. 11, no. 1, pp. 50–62, Jan. 2024, doi: 10.25126/jtiik.2024118315.
L. W. Astuti, Y. Sari, and Suprapto, “Code-mixed sentiment analysis using transformer for Twitter social media data,” Int. J. Adv. Comput. Sci. Appl., vol. 14, no. 10, pp. 1–8, Oct. 2023, doi: 10.14569/IJACSA.2023.0141053.
T. D. Purnomo and J. Sutopo, “Comparison of pre-trained BERT-based transformer models for regional language text sentiment analysis in Indonesia,” Int. J. Sci. Technol., vol. 3, no. 3, pp. 1–9, Jul. 2024, doi: 10.56127/ijst.v3i3.1739.
M. F. Mubaraq and W. Maharani, “Sentiment analysis on Twitter social media towards climate change in Indonesia using IndoBERT model,” J. Media Inform. Budidarma, vol. 6, no. 4, pp. 2009–2018, Dec. 2022, doi: 10.30865/mib.v6i4.4570.
R. Perwira, V. A. Permadi, D. I. Purnamasari, and R. P. Agusdin, “Domain-specific fine-tuning of IndoBERT for aspect-based sentiment analysis in Indonesian travel user-generated content,” J. Inf. Syst. Eng. Bus. Intell., vol. 11, no. 1, pp. 30–40, Jan. 2025, doi: 10.20473/jisebi.11.1.30-40.
D. Febrianto, M. A. Fitriani, M. Afrad, and M. A. Khadija, “Aspect-based sentiment analysis menggunakan IndoBERT model terhadap review pengunjung objek wisata Baturraden,” Melek IT: Inf. Technol. J., vol. 10, no. 2, pp. 150–160, Jun. 2024, doi: 10.30742/melekitjournal.v10i2.358.
D. Dhendra and V. G. Utomo, “Benchmarking IndoBERT and transformer models for sentiment classification on Indonesian e-government service reviews,” J. Transformatika, vol. 23, no. 1, pp. 1–9, Jan. 2025, doi: 10.26623/transformatika.v23i1.12095.
Y. Aunugu, “The role of AI in customer sentiment analysis for strategic decisions,” Int. J. Comput. Sci. Rev. Res. (IJCSRR), vol. 8, no. 3, pp. 552–563, Mar. 2025, doi: 10.5281/zenodo.11283047.
A. Amrullah, “Sentiment analysis in the age of transformers and large language models: Future directions,” Intell. Comput. J., vol. 7, no. 2, pp. 33–44, Apr. 2025, doi: 10.56789/icj.v7i2.9020.
Y. Abdelwahab, M. Kholief, and A. Sedky, “Justifying Arabic text sentiment analysis using explainable AI (XAI): LASIK surgeries case study,” Information, vol. 13, no. 11, pp. 1–12, Nov. 2022, doi: 10.3390/info13110536.
F. Jourdan, “Advancing fairness in natural language processing: The role of explainability in model design,” Ph.D. thesis, ETH Zürich, Zürich, Switzerland, 2024, doi: 10.3929/ethz-b-000662511.
I. Villanueva-Miranda, Y. Xie, and G. Xiao, “Sentiment analysis in public health: A systematic review of the current state, challenges, and future directions,” Front. Public Health, vol. 13, no. 4, pp. 1–15, Mar. 2025, doi: 10.3389/fpubh.2025.1609749.
A. Maretta and A. Meiriza, “Aspect-based sentiment analysis of hospital service reviews using fine-tuned IndoBERT,” J. Appl. Informatics Comput., vol. 9, no. 5, pp. 101–110, Oct. 2025, doi: 10.30871/jaic.v9i5.10765.
N. Lin, Y. Fu, X. Lin, D. Zhou, A. Yang, and S. Jiang, “CL-XABSA: Contrastive learning for cross-lingual aspect-based sentiment analysis,” IEEE/ACM Trans. Audio Speech Lang. Process., vol. 31, no. Dec., pp. 2935–2946, Dec. 2022, doi: 10.1109/TASLP.2023.3297964.
V. R. Prasetyo, M. F. Naufal, and K. Wijaya, “Sentiment analysis of ChatGPT on Indonesian text using hybrid CNN and Bi-LSTM,” J. RESTI (Rekayasa Sistem dan Teknologi Informasi), vol. 9, no. 2, pp. 327–333, Apr. 2025, doi: 10.29207/resti.v9i2.6334.
P. Cristin, B. Natalia, J. C. Limantara, and Sarwosri, “Performance comparison of embeddings and keyword selection methods in enterprise documents,” J. Appl. Informatics Comput., vol. 9, no. 4, pp. 112–118, Sep. 2025, doi: 10.30871/jaic.v9i4.9971.
F. Koto, A. Rahimi, J. H. Lau, and T. Baldwin, “IndoLEM and IndoBERT: A benchmark dataset and pre-trained language model for Indonesian NLP,” Comput. Linguist. Conf. (COLING), vol. 2020, no. Dec., pp. 757–770, Dec. 2020, doi: 10.18653/v1/2020.coling-main.66.
Bila bermanfaat silahkan share artikel ini
Berikan Komentar Anda terhadap artikel Explainable Aspect-Based Sentiment Analysis with Contrast-Aware IndoBERT for Indonesian Public Service Reviews
Pages: 2229-2239
Copyright (c) 2026 Muhammad Shihab Fathurrahman Jondien, Taqwa Hariguna, Dhanar Intan Surya Saputra
This work is licensed under a Creative Commons Attribution 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 Creative Commons Attribution 4.0 International 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 (Refer to The Effect of Open Access).
