Elamur Research⌕
☰
home / papers / 2026 / From Pain to Praise: Aspect-Based Sentiment Analysis for Norwegian Patient Feedback

From Pain to Praise: Aspect-Based Sentiment Analysis for Norwegian Patient Feedback

Paper · 2026 · Lilja Charlotte Storset, Elma Jelin, Rebecka Maria Norman et al.

Summary

Norway's public health institute (NIPH) collects tens of thousands of free-text comments each year through national patient experience surveys. Because staff currently analyse only a small manual sample, most of this feedback goes unused. The authors address this by introducing NorPaC absa, an extension of the Norwegian Patient Comment corpus and the first formal aspect-based sentiment analysis (ABSA) resource for Norwegian. Healthcare experts labelled 2382 comments about general practitioners (GP) and specialist mental healthcare (SMH). Labels cover 25 fine-grained aspect categories, each with a polarity, at both sentence and comment level.

The team then compared:

  • Architectures: a fine-tuned encoder (NorBERT3 large) against a decoder (NorMistral-11b-thinking), including prompting setups.
  • Task design: joint versus sequential prediction of aspects and polarity.
  • Granularity: different text and label levels.

The smaller encoder came out ahead. Training on sentences and aggregating the predictions per comment worked better than training on whole comments. Predicting combined aspect–polarity labels in one step proved viable. This work points toward scalable automated analysis of patient feedback that could inform service improvement.

Key points

  • NorPaC absa has 2382 comments (1242 GP, 1140 SMH) and 5563 annotated aspects, averaging about 2.3 aspects per comment, so the task is treated as multi-label.
  • The 25 aspect categories include 11 shared across both domains. A "no aspect" label marks neutral sentences that lack a polar target.
  • Polarity is mostly positive (45.8%) or negative (43.3%), with some mixed (9.3%) and very few neutral (1.7%) cases.
  • NorBERT3 large beat NorMistral-11b-thinking. Fine-tuned models tended to over-predict frequent labels such as "general" and "patient-provider/staff relationships".
  • The data itself stays private for privacy reasons. The guidelines, code and models are published at github.com/ltgoslo/From_Pain_to_Praise.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Institutions

🎓University of OsloOslo · Norway51 works · 4,155 citations

Also on the byline: Norwegian Institute of Public Health.

 Fields, topics and concepts

📚 Shelves 6The field of science this work belongs to (OpenAlex fields).Computer Science · 18,614 worksArtificial Intelligence · 17,965 worksArts and Humanities · 17,450 worksLanguage and Linguistics · 17,296 worksHealth Professions · 75 worksGeneral Health Professions · 14 works

🧭 Topics 1Research subjects and languages the work is about.Sentiment & emotion · 1,035 works

Cite

Cite as Lilja Charlotte Storset, Elma Jelin, Rebecka Maria Norman, Oyvind Bjertnaes, Lilja Øvrelid and Erik Velldal. 2026. From Pain to Praise: Aspect-Based Sentiment Analysis for Norwegian Patient Feedback. Proceedings of the 1st Workshop on Linguistic Analysis for Health (HeaLing 2026).

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2026.healing-1.16

@inproceedings{storset2026pain,
  title = {{From Pain to Praise: Aspect-Based Sentiment Analysis for Norwegian Patient Feedback}},
  author = {Lilja Charlotte Storset, Elma Jelin, Rebecka Maria Norman, Oyvind Bjertnaes, Lilja Øvrelid and Erik Velldal},
  booktitle = {Proceedings of the 1st Workshop on Linguistic Analysis for Health (HeaLing 2026)},
  year = {2026},
  doi = {10.18653/v1/2026.healing-1.16},
  url = {https://doi.org/10.18653/v1/2026.healing-1.16}
}

References

20 references, 5 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Muhammad Abdul-Mageed, AbdelRahim Elmadany, and El Moatez Billah Nagoudi. 2021. ARBERT & MARBERT: Deep Bidirectional Transformers for Arabic. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers), pages 7088–7105, Online. Association for Computational Linguistics. ↗
  2. Omer S. Alkhnbashi, Rasheed Mohammad, and Mohammad Hammoudeh. 2024. Aspect-Based Sentiment Analysis of Patient Feedback Using Large Language Models. Big Data and Cognitive Computing, 8(12):167. ↗
  3. Seba Alnasser and Sarab Almuhaideb. 2024. Listening to Patients: Advanced Arabic Aspect-Based Sentiment Analysis Using Transformer Models Towards Better Healthcare. Big Data and Cognitive Computing, 8:156. ↗
  4. Rachel Grob, Mark Schlesinger, Lacey Rose Barre, Naomi Bardach, Tara Lagu, Dale Shaller, Andrew M. Parker, Steven C. Martino, Melissa L. Finucane, Jennifer L. Cerully, and Alina Palimaru. 2019. What Words Convey: The Potential for Patient Narratives to Inform Quality Improvement. The Milbank Quarterly, 97(1):176–227. ↗
  5. Hsiu-Fang Hsieh and Sarah E. Shannon. 2005. Three approaches to qualitative content analysis. Qualitative Health Research, 15(9):1277–1288. ↗
  6. Elma Jelin, Lilja Charlotte Storset, Rebecka M Norman, Hilde Hestad Hestad Iversen, Lina Harvold Ellingsen-Dalskau, Petter Mæhlum, Erik Velldal, Lilja Øvrelid, and Oyvind Bjertnaes. 2025. From words to action? A scoping review on automatic sentiment analysis of patient experience comments from online sources and surveys. BMJ Health & Care Informatics, 32(1):e101631. find
  7. Bing Liu. 2015. Sentiment Analysis: Mining Opinions, Sentiments, and Emotions. Cambridge University Press, Cambridge. find
  8. Petter Mæhlum, David Samuel, Rebecka Maria Norman, Elma Jelin, Øyvind Andresen Bjertnæs, Lilja Øvrelid, and Erik Velldal. 2024. It’s Difficult to Be Neutral – Human and LLM-based Sentiment Annotation of Patient Comments. In Proceedings of the First Workshop on Patient-Oriented Language Processing (CL4Health) @ LREC-COLING 2024, pages 8–19, Torino, Italia. ELRA and ICCL. find
  9. Ambreen Nazir, Yuan Rao, Lianwei Wu, and Ling Sun. 2022. Issues and Challenges of Aspect-based Sentiment Analysis: A Comprehensive Survey. IEEE Transactions on Affective Computing, 13(2):845–863. find
  10. Rebecka Maria Norman, Elma Jelin, and Oyvind Bjertnaes. 2024. Multimorbidity and patient experience with general practice: A national cross-sectional survey in Norway. BMC Primary Care, 25(1):249. ↗
  11. Maria Pontiki, Dimitris Galanis, Haris Papageorgiou, Ion Androutsopoulos, Suresh Manandhar, Mohammad AL-Smadi, Mahmoud Al-Ayyoub, Yanyan Zhao, Bing Qin, Orphee De Clercq, Veronique Hoste, Marianna Apidianaki, Xavier Tannier, Natalia Loukachevitch, Evgeniy Kotelnikov, Núria Bel, Salud María Jiménez-Zafra, and Gülşen Eryiğit. 2016. SemEval-2016 Task 5: Aspect Based Sentiment Analysis. In Proceedings of the 10th International Workshop on Semantic Evaluation (SemEval-2016), pages 19–30, San Diego, California. Association for Computational Linguistics. ↗
  12. Wenxuan Zhang, Xin Li, Yang Deng, Lidong Bing, and Wai Lam. 2023. A Survey on Aspect-Based Sentiment Analysis: Tasks, Methods, and Challenges. IEEE Transactions on Knowledge and Data Engineering, 35(11):11019–11038. ↗
  13. Maria Pontiki, Dimitris Galanis, Haris Papageorgiou, Suresh Manandhar, and Ion Androutsopoulos. 2015. SemEval-2015 Task 12: Aspect Based Sentiment Analysis. In Proceedings of the 9th International Workshop on Semantic Evaluation (SemEval 2015), pages 486–495, Denver, Colorado. Association for Computational Linguistics.
  14. Maria Pontiki, Dimitris Galanis, John Pavlopoulos, Harris Papageorgiou, Ion Androutsopoulos, and Suresh Manandhar. 2014. SemEval-2014 Task 4: Aspect Based Sentiment Analysis. In Proceedings of the 8th International Workshop on Semantic Evaluation (SemEval 2014), pages 27–35, Dublin, Ireland. Association for Computational Linguistics.
  15. Erik Riiskjær, Jette Ammentorp, and Poul-Erik Kofoed. 2012. The value of open-ended questions in surveys on patient experience: number of comments and perceived usefulness from a hospital perspective. International Journal for Quality in Health Care: Journal of the International Society for Quality in Health Care, 24(5):509–516. ↗
  16. Carol Rivas, Daria Tkacz, Laurence Antao, Emmanouil Mentzakis, Margaret Gordon, Sydney Anstee, and Richard Giordano. 2019. Automated analysis of free-text comments and dashboard representations in patient experience surveys: a multimethod co-design study. Health Services and Delivery Research. NIHR Journals Library, Southampton (UK). ↗
  17. Egil Rønningstad, Lilja Charlotte Storset, Petter Mæhlum, Lilja Øvrelid, and Erik Velldal. 2025. Mixed Feelings: Cross-Domain Sentiment Classification of Patient Feedback. In Proceedings of the Joint 25th Nordic Conference on Computational Linguistics and 11th Baltic Conference on Human Language Technologies (NoDaLiDa/Baltic-HLT 2025), pages 537– 543, Tallinn, Estonia. University of Tartu Library. ↗
  18. David Samuel, Lilja Øvrelid, Erik Velldal, and Andrey Kutuzov. 2025. Fluent Alignment with Disfluent Judges: Post-training for Lower-resource Languages. In Proceedings of the Fourteenth International Conference of Learning Representations (ICLR 2026), Rio de Janeiro, Brazil. ↗
  19. Paul F. Simmering and Paavo Huoviala. 2023. Large language models for aspect-based sentiment analysis. arXiv preprint. ArXiv:2310.18025 [cs]. ↗
  20. Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Łukasz Kaiser, and Illia Polosukhin. 2017. Attention is all you need. Advances in neural information processing systems, 30.

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/storset-2026-pain-praise-aspect-based-sentiment/