---
title: "A Synthetic Conversational Dataset for Type 2 Diabetes"
description: "Real conversations between patients and clinicians are hard to obtain because of privacy rules. 30 references."
authors: ["Stergios Ntanavaras", "Maaike de Boer", "Piek T.J.M. Vossen"]
year: 2026
venue: "Proceedings of the Third Workshop on Patient-Oriented Language Processing (CL4Health) @ LREC 2026"
source: "https://aclanthology.org/2026.cl4health-1.16/"
doi: "10.63317/3cbhekpxj33y"
licence: "CC BY-NC 4.0 (ELRA, LREC proceedings)"
licence_url: "https://creativecommons.org/licenses/by-nc/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/ntanavaras-2026-synthetic-conversational-dataset-type-2/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/ntanavaras-2026-synthetic-conversational-dataset-type-2/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2026](https://md.research.elamur.ai/free/en/papers/2026/page/20/) / A Synthetic Conversational Dataset for Type 2 Diabetes Management

# A Synthetic Conversational Dataset for Type 2 Diabetes Management

**Authors:** Stergios Ntanavaras, [Maaike de Boer](https://md.research.elamur.ai/free/en/authors/maaike-de-boer/) and [Piek T.J.M. Vossen](https://md.research.elamur.ai/free/en/authors/piek-vossen/) · **Year:** 2026 · **Venue:** [Proceedings of the Third Workshop on Patient-Oriented Language Processing (CL4Health) @ LREC 2026](https://md.research.elamur.ai/free/en/venues/cl4health/) ([Workshop on Patient-Oriented Language Processing @ LREC 2026](https://md.research.elamur.ai/free/en/venues/cl4health/2026/)) · **Publisher:** [European Language Resources Association (ELRA)](https://md.research.elamur.ai/free/en/publishers/european-language-resources-association-elra/) · **Licence:** [CC BY-NC 4.0 (ELRA, LREC proceedings)](https://creativecommons.org/licenses/by-nc/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/2026.cl4health-1.16/) · [PDF](https://aclanthology.org/2026.cl4health-1.16.pdf) · [DOI 10.63317/3cbhekpxj33y](https://doi.org/10.63317/3cbhekpxj33y) · [OpenAlex](https://openalex.org/W7166576389)

**Citations:** 0 citations (OpenAlex, 2026-10-14)

## Summary
Real conversations between patients and clinicians are hard to obtain because of privacy rules. This leaves few annotated dialogue resources for chronic conditions such as Type 2 Diabetes, and existing ones mostly cover human–human talk rather than patient–agent interaction. The authors address this by building a fully synthetic dataset for conversational triple extraction. They used GPT-4 with structured, instruction-rich prompts. First, they created 16 patient personas based on Dutch epidemiological and demographic statistics. Next, they generated 256 multi-turn dialogues between these personas and a caretaker agent. Finally, they labeled the dialogue tokens with Subject–Predicate–Object (SPO) roles, combining manual work with LLM-assisted annotation. Both LLM-based judges and human experts assessed the conversations. According to these evaluations, the dialogues show personalization, empathy, contextual awareness and medically sound advice. The resource feeds the knowledge-extraction module of the CHIP hybrid intelligence platform, where extracted triples populate a User Knowledge Graph for personalized lifestyle support. The work shows that generative models can produce realistic healthcare dialogue at scale for data-poor, privacy-sensitive domains.

## Key points
- **Dataset size and structure:** 16 diabetic personas and 256 synthetic patient–agent conversations. Each conversation has 4–6 exchanges, and openings are split roughly evenly between patient-initiated and agent-initiated.
- **Persona design:** Personas draw on Dutch sources, including the DPARD registry (mean outpatient age 65.9) and CBS StatLine.
  - Ages are 55, 60, 65, 70 and 75.
  - Ethnicities are Dutch, Moroccan, Surinamese and Turkish. These were chosen because diabetes prevalence is higher in the Moroccan (7.09%), Surinamese (6.15%) and Turkish (5.95%) groups than among Dutch individuals (1.99%).
  - Each persona has a culturally typical name, and gender and ethnicity are balanced.
- **Annotation:** Token-level SPO labels are produced through a hybrid of manual and LLM-automated annotation. They are intended for training and benchmarking triple-extraction models.
- **Project context:** The dataset is part of the Dutch CHIP initiative, a modular microservice platform that combines dialogue processing, information extraction, a User Knowledge Graph, a Domain Knowledge Graph and reasoning.
- **Availability:** Personas, conversations and annotations are released publicly on GitHub (stergios97/synthetic-conversational-data-diabetes-management).

*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

- [Vrije Universiteit Amsterdam](https://md.research.elamur.ai/free/en/universities/vrije-universiteit-amsterdam/) — University, Amsterdam, Netherlands · 42 works · 3,372 citations
- [Netherlands Organisation for Applied Scientific Research](https://md.research.elamur.ai/free/en/labs/netherlands-organisation-for-applied-scientific-research/) — Research institute, Delft, Netherlands · 2 works

📚 Shelves (6): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 16,667 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 16,194 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 16,039 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 15,891 works · [🩺 Medicine](https://md.research.elamur.ai/free/en/shelves/medicine/) · 241 works · [🩺 Health Informatics](https://md.research.elamur.ai/free/en/shelves/medicine/health-informatics/) · 184 works

🧭 Topics (1): [🗨️ Dialogue systems](https://md.research.elamur.ai/free/en/topics/dialogue-systems/) · 624 works

## Cite

Cite as: Stergios Ntanavaras, Maaike de Boer and Piek T.J.M. Vossen. 2026. A Synthetic Conversational Dataset for Type 2 Diabetes Management. Proceedings of the Third Workshop on Patient-Oriented Language Processing (CL4Health) @ LREC 2026.

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@inproceedings{ntanavaras2026synthetic,
  title = {{A Synthetic Conversational Dataset for Type 2 Diabetes Management}},
  author = {Stergios Ntanavaras, Maaike {de Boer} and Piek T.J.M. Vossen},
  booktitle = {Proceedings of the Third Workshop on Patient-Oriented Language Processing (CL4Health) @ LREC 2026},
  year = {2026},
  doi = {10.63317/3cbhekpxj33y},
  url = {https://doi.org/10.63317/3cbhekpxj33y}
}
```

- [← previous A Syntactic and Semantic Probe into Language Evolution based on Large Language Models](https://md.research.elamur.ai/free/en/papers/pang-2026-syntactic-semantic-probe-into-language/)
- [next → A Systematic Exploration of Text Decomposition and Budget Distribution in Differentially Private Text Obfuscation](https://md.research.elamur.ai/free/en/papers/meisenbacher-2026-systematic-exploration-text-decomposition-budget/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/20/)

## References

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

1. Jessica C. G. Bak, Dick Mul, Erik H. Serné, Harold W. de Valk, Theo C. J. Sas, Petronella H. Geelhoed-Duijvestijn, Mark H. H. Kramer, Max Nieuwdorp, and Carianne L. Verheugt. 2021. Dpard: rationale, design and initial results from the dutch national diabetes registry. BMC Endocrine Disorders, 21(1):… [find](https://scholar.google.com/scholar?q=Jessica+C%2E+G%2E+Bak%2C+Dick+Mul%2C+Erik+H%2E+Sern%C3%A9%2C+Harold+W%2E+de+Valk%2C+Theo+C%2E+J%2E+Sas%2C+Petronella+H%2E+Geelhoed%2DDuijvestijn%2C+Mark+H%2E+H%2E+Kramer%2C+Max+Nieuwdorp%2C+and+Carianne+L%2E+Verheugt%2E+2021%2E+Dpard%3A+rationale%2C+d)
2. Centraal Bureau voor de Statistiek. 2024. Gezond-heid en zorggebruik; persoonskenmerken. CBS StatLine. Accessed: 2024-06-02. [find](https://scholar.google.com/scholar?q=Centraal+Bureau+voor+de+Statistiek%2E+2024%2E+Gezond%2Dheid+en+zorggebruik%3B+persoonskenmerken%2E+CBS+StatLine%2E+Accessed%3A+2024%2D06%2D02%2E)
3. Maryam Changizi and Mohammad H. Kaveh. 2017. Effectiveness of the mhealth technology in improvement of healthy behaviors in an elderly population—a systematic review. mHealth, 3(11). [find](https://scholar.google.com/scholar?q=Maryam+Changizi+and+Mohammad+H%2E+Kaveh%2E+2017%2E+Effectiveness+of+the+mhealth+technology+in+improvement+of+healthy+behaviors+in+an+elderly+population%E2%80%94a+systematic+review%2E+mHealth%2C+3%2811%29%2E)
4. Anshul Chavda and Pushpak Bhattacharyya. Synthetic dialogue data generation: A comprehensive survey of methods, evaluation, and challenges. [find](https://scholar.google.com/scholar?q=Anshul+Chavda+and+Pushpak+Bhattacharyya%2E+Synthetic+dialogue+data+generation%3A+A+comprehensive+survey+of+methods%2C+evaluation%2C+and+challenges%2E)
5. Floris den Hengst, Shaad Alaka, and Bart A. Kamphorst. 2026. Collaborative hybrid intelligence platform chip: A modular architecture for developing and testing personalized lifestyle support interactions. SoftwareX, 33:102536. [find](https://scholar.google.com/scholar?q=Floris+den+Hengst%2C+Shaad+Alaka%2C+and+Bart+A%2E+Kamphorst%2E+2026%2E+Collaborative+hybrid+intelligence+platform+chip%3A+A+modular+architecture+for+developing+and+testing+personalized+lifestyle+support+interacti)
6. Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2019. [Bert: Pre-training of deep bidirectional transformers for language understanding](https://md.research.elamur.ai/free/en/papers/devlin-2019-bert-pre-training-deep-bidirectional/). In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies,…
7. Bernd JW Dudzik, Jasper S van der Waa, Pei-Yu Chen, Roel Dobbe, Ínigo MDR de Troya, Roos M Bakker, Maaike HT de Boer, Quirine TS Smit, Davide Dell’Anna, Emre Erdogan, et al. 2024. Hybrid intelligence supports application development for diabetes lifestyle management. Journal of Artificial Intelligen… [find](https://scholar.google.com/scholar?q=Bernd+JW+Dudzik%2C+Jasper+S+van+der+Waa%2C+Pei%2DYu+Chen%2C+Roel+Dobbe%2C+%C3%8Dnigo+MDR+de+Troya%2C+Roos+M+Bakker%2C+Maaike+HT+de+Boer%2C+Quirine+TS+Smit%2C+Davide+Dell%E2%80%99Anna%2C+Emre+Erdogan%2C+et+al%2E+2024%2E+Hybrid+intelligence+)
8. Kilem L Gwet. 2014. Handbook of inter-rater reliability: The definitive guide to measuring the extent of agreement among raters. Advanced Analytics, LLC. [find](https://scholar.google.com/scholar?q=Kilem+L+Gwet%2E+2014%2E+Handbook+of+inter%2Drater+reliability%3A+The+definitive+guide+to+measuring+the+extent+of+agreement+among+raters%2E+Advanced+Analytics%2C+LLC%2E)
9. Samuel Holmes, Anne Moorhead, Raymond Bond, Huiru Zheng, Vivien Coates, and Michael McTear. 2019. Usability testing of a healthcare chatbot: Can we use conventional methods to assess conversational user interfaces? In Proceedings of the 31st European conference on cognitive ergonomics, pages 207–214… [find](https://scholar.google.com/scholar?q=Samuel+Holmes%2C+Anne+Moorhead%2C+Raymond+Bond%2C+Huiru+Zheng%2C+Vivien+Coates%2C+and+Michael+McTear%2E+2019%2E+Usability+testing+of+a+healthcare+chatbot%3A+Can+we+use+conventional+methods+to+assess+conversational+us)
10. Or Honovich, Thomas Scialom, Omer Levy, and Timo Schick. 2022. [Unnatural instructions: Tuning language models with (almost) no human labor](https://md.research.elamur.ai/free/en/papers/honovich-2023-unnatural-instructions-tuning-language-models/). arXiv preprint arXiv:2212.09689.
11. Jan-Christoph Klie, Michael Bugert, Beto Boullosa, Richard Eckart de Castilho, and Iryna Gurevych. 2018. [The INCEpTION platform: Machine-assisted and knowledge-oriented interactive annotation](https://md.research.elamur.ai/free/en/papers/klie-2018-inception-platform-machine-assisted-knowledge/). In Proceedings of the 27th International Conference on Computational Linguistics: System Demonstrations, pa…
12. Rensis Likert. 1932. A technique for the measurement of attitudes. Archives of Psychology, 22(140):1–55. [find](https://scholar.google.com/scholar?q=Rensis+Likert%2E+1932%2E+A+technique+for+the+measurement+of+attitudes%2E+Archives+of+Psychology%2C+22%28140%29%3A1%E2%80%9355%2E)
13. Yang Liu, Dan Iter, Yichong Xu, Shuohang Wang, Ruochen Xu, and Chenguang Zhu. 2023. [G-eval: Nlg evaluation using gpt-4 with better human alignment](https://md.research.elamur.ai/free/en/papers/liu-2023-g-eval-nlg-evaluation-gpt/).
14. OpenAI. 2024. [Gpt-4 technical report](https://md.research.elamur.ai/free/en/papers/openai-2023-gpt-4-technical-report/).
15. Arjun Panickssery, Samuel R. Bowman, and Shi Feng. 2024. Llm evaluators recognize and favor their own generations. In Advances in Neural Information Processing Systems, volume 37, pages 68772–68802. Curran Associates, Inc. [find](https://scholar.google.com/scholar?q=Arjun+Panickssery%2C+Samuel+R%2E+Bowman%2C+and+Shi+Feng%2E+2024%2E+Llm+evaluators+recognize+and+favor+their+own+generations%2E+In+Advances+in+Neural+Information+Processing+Systems%2C+volume+37%2C+pages+68772%E2%80%9368802%2E+C)
16. Tessa F Peerbolte, Rozanne JA van Diggelen, Pieter van den Haak, Kim Geurts, Luc JW Evers, Bastiaan R Bloem, Nienke M de Vries, and Sanne W van den Berg. 2025. Conversational agents supporting self-management in people with a chronic disease: Systematic review. J Med Internet Res, 27:e72309. [find](https://scholar.google.com/scholar?q=Tessa+F+Peerbolte%2C+Rozanne+JA+van+Diggelen%2C+Pieter+van+den+Haak%2C+Kim+Geurts%2C+Luc+JW+Evers%2C+Bastiaan+R+Bloem%2C+Nienke+M+de+Vries%2C+and+Sanne+W+van+den+Berg%2E+2025%2E+Conversational+agents+supporting+self%2Dma)
17. Krithika Ramesh, Nupoor Gandhi, Pulkit Madaan, Lisa Bauer, Charith Peris, and Anjalie Field. 2024. Evaluating differentially private synthetic data generation in high-stakes domains. arXiv preprint arXiv:2410.08327. [find](https://scholar.google.com/scholar?q=Krithika+Ramesh%2C+Nupoor+Gandhi%2C+Pulkit+Madaan%2C+Lisa+Bauer%2C+Charith+Peris%2C+and+Anjalie+Field%2E+2024%2E+Evaluating+differentially+private+synthetic+data+generation+in+high%2Dstakes+domains%2E+arXiv+preprint+ar)
18. Selene Báez Santamarıa, Lea Krause, Lucia Donatelli, and Piek Vossen. 2023. The role of personal perspectives in open-domain dialogue. [find](https://scholar.google.com/scholar?q=Selene+B%C3%A1ez+Santamar%C4%B1a%2C+Lea+Krause%2C+Lucia+Donatelli%2C+and+Piek+Vossen%2E+2023%2E+The+role+of+personal+perspectives+in+open%2Ddomain+dialogue%2E)
19. Heydar Soudani, Evangelos Kanoulas, and Faegheh Hasibi. 2023. Data augmentation for conversational ai. In Proceedings of the 32nd ACM International Conference on Information and Knowledge Management, pages 5220– 5223. [find](https://scholar.google.com/scholar?q=Heydar+Soudani%2C+Evangelos+Kanoulas%2C+and+Faegheh+Hasibi%2E+2023%2E+Data+augmentation+for+conversational+ai%2E+In+Proceedings+of+the+32nd+ACM+International+Conference+on+Information+and+Knowledge+Management%2C+)
20. Joanne K. Ujcic-Voortman, Miranda T. Schram, Monique A. Jacobs-van der Bruggen, Arnoud P. Verhoeff, and Caroline A. Baan. 2009. Diabetes prevalence and risk factors among ethnic minorities. European Journal of Public Health, 19(5):511–515. [find](https://scholar.google.com/scholar?q=Joanne+K%2E+Ujcic%2DVoortman%2C+Miranda+T%2E+Schram%2C+Monique+A%2E+Jacobs%2Dvan+der+Bruggen%2C+Arnoud+P%2E+Verhoeff%2C+and+Caroline+A%2E+Baan%2E+2009%2E+Diabetes+prevalence+and+risk+factors+among+ethnic+minorities%2E+European+J)
21. Klaus Krippendorff. 2011. Computing krippendorff’s alpha-reliability. [find](https://scholar.google.com/scholar?q=Klaus+Krippendorff%2E+2011%2E+Computing+krippendorff%E2%80%99s+alpha%2Dreliability%2E)
22. E H Wagner, B T Austin, and M Von Korff. 1996. Organizing care for patients with chronic illness. The Milbank Quarterly, 74(4):511–544. [find](https://scholar.google.com/scholar?q=E+H+Wagner%2C+B+T+Austin%2C+and+M+Von+Korff%2E+1996%2E+Organizing+care+for+patients+with+chronic+illness%2E+The+Milbank+Quarterly%2C+74%284%29%3A511%E2%80%93544%2E)
23. JR Landis. 1977. The measurement of observer agreement for categorical data. Biometrics. 11. [find](https://scholar.google.com/scholar?q=JR+Landis%2E+1977%2E+The+measurement+of+observer+agreement+for+categorical+data%2E+Biometrics%2E+11%2E)
24. Liliana Laranjo, Adam G Dunn, Huong Ly Tong, Ahmet Baki Kocaballi, Jessica Chen, Rabia Bashir, Didi Surian, Blanca Gallego, Farah Magrabi, Annie YS Lau, et al. 2018. Conversational agents in healthcare: a systematic review. Journal of Steven Bedrick and A. Seza Doğruöz and Sergiu Nisioi. 2025. A Typ… [find](https://scholar.google.com/scholar?q=Liliana+Laranjo%2C+Adam+G+Dunn%2C+Huong+Ly+Tong%2C+Ahmet+Baki+Kocaballi%2C+Jessica+Chen%2C+Rabia+Bashir%2C+Didi+Surian%2C+Blanca+Gallego%2C+Farah+Magrabi%2C+Annie+YS+Lau%2C+et+al%2E+2018%2E+Conversational+agents+in+healthcar)
25. Language Resource References Trisha Das and Dina Albassam and Jimeng Sun. 2024. Synthetic Patient-Physician Dialogue Generation from Clinical Notes Using LLM. [find](https://scholar.google.com/scholar?q=Language+Resource+References+Trisha+Das+and+Dina+Albassam+and+Jimeng+Sun%2E+2024%2E+Synthetic+Patient%2DPhysician+Dialogue+Generation+from+Clinical+Notes+Using+LLM%2E)
26. Demszky, Dorottya and Movshovitz-Attias, Dana and Ko, Jeongwoo and Cowen, Alan and Nemade, Gaurav and Ravi, Sujith. 2020. GoEmotions: A dataset of fine-grained emotions. [find](https://scholar.google.com/scholar?q=Demszky%2C+Dorottya+and+Movshovitz%2DAttias%2C+Dana+and+Ko%2C+Jeongwoo+and+Cowen%2C+Alan+and+Nemade%2C+Gaurav+and+Ravi%2C+Sujith%2E+2020%2E+GoEmotions%3A+A+dataset+of+fine%2Dgrained+emotions%2E)
27. Eisenberg, Joshua and Sheriff, Michael. 2020. Automatic extraction of personal events from dialogue. [find](https://scholar.google.com/scholar?q=Eisenberg%2C+Joshua+and+Sheriff%2C+Michael%2E+2020%2E+Automatic+extraction+of+personal+events+from+dialogue%2E)
28. Ahmad Rezaie Mianroodi and Amirali Rezaie and Niko Grisel Todorov and Cyril Rakovski and Frank Rudzicz. 2025. MedSynth: Realistic, Synthetic Medical Dialogue-Note Pairs. [find](https://scholar.google.com/scholar?q=Ahmad+Rezaie+Mianroodi+and+Amirali+Rezaie+and+Niko+Grisel+Todorov+and+Cyril+Rakovski+and+Frank+Rudzicz%2E+2025%2E+MedSynth%3A+Realistic%2C+Synthetic+Medical+Dialogue%2DNote+Pairs%2E)
29. Wang, Junda and Yao, Zonghai and Yang, Zhichao and Zhou, Huixue and Li, Rumeng and Wang, Xun and Xu, Yucheng and Yu, Hong. 2024. NoteChat: A Dataset of Synthetic Patient-Physician Conversations Conditioned on Clinical Notes. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Wang%2C+Junda+and+Yao%2C+Zonghai+and+Yang%2C+Zhichao+and+Zhou%2C+Huixue+and+Li%2C+Rumeng+and+Wang%2C+Xun+and+Xu%2C+Yucheng+and+Yu%2C+Hong%2E+2024%2E+NoteChat%3A+A+Dataset+of+Synthetic+Patient%2DPhysician+Conversations+Condit)
30. Yu, Dian and Yu, Zhou. 2019. Midas: A dialog act annotation scheme for open domain human machine spoken conversations. [find](https://scholar.google.com/scholar?q=Yu%2C+Dian+and+Yu%2C+Zhou%2E+2019%2E+Midas%3A+A+dialog+act+annotation+scheme+for+open+domain+human+machine+spoken+conversations%2E)

*Updated 1 November 2026*