IDEAS home Printed from https://ideas.repec.org/a/gam/jftint/v17y2025i4p138-d1618786.html
   My bibliography  Save this article

Benchmarking Large Language Models from Open and Closed Source Models to Apply Data Annotation for Free-Text Criteria in Healthcare

Author

Listed:
  • Ali Nemati

    (Health Informatics Department, Zilber College of Public Health, University of Wisconsin, Milwaukee, WI 53211, USA
    These authors contributed equally to this work.)

  • Mohammad Assadi Shalmani

    (Health Informatics Department, Zilber College of Public Health, University of Wisconsin, Milwaukee, WI 53211, USA
    These authors contributed equally to this work.)

  • Qiang Lu

    (Beijing Key Laboratory of Petroleum Data Mining, China University of Petroleum, Beijing 102249, China)

  • Jake Luo

    (Health Informatics & Administration Department, Zilber College of Public Health, University of Wisconsin, Milwaukee, WI 53211, USA)

Abstract

Large language models (LLMs) hold the potential to significantly enhance data annotation for free-text healthcare records. However, ensuring their accuracy and reliability is critical, especially in clinical research applications requiring the extraction of patient characteristics. This study introduces a novel evaluation framework based on Multi-Criteria Decision Analysis (MCDA) and the Order of Preference by Similarity to Ideal Solution (TOPSIS) technique, designed to benchmark LLMs on their annotation quality. The framework defines ten evaluation metrics across key criteria such as age, gender, BMI, disease presence, and blood markers (e.g., white blood count and platelets). Using this methodology, we assessed leading open source and commercial LLMs, achieving accuracy scores of 0.59, 1, 0.84, 0.56, and 0.92, respectively, for the specified criteria. Our work not only provides a rigorous framework for evaluating LLM capabilities in healthcare data annotation but also highlights their current performance limitations and strengths. By offering a comprehensive benchmarking approach, we aim to support responsible adoption and decision-making in healthcare applications.

Suggested Citation

  • Ali Nemati & Mohammad Assadi Shalmani & Qiang Lu & Jake Luo, 2025. "Benchmarking Large Language Models from Open and Closed Source Models to Apply Data Annotation for Free-Text Criteria in Healthcare," Future Internet, MDPI, vol. 17(4), pages 1-27, March.
  • Handle: RePEc:gam:jftint:v:17:y:2025:i:4:p:138-:d:1618786
    as

    Download full text from publisher

    File URL: https://www.mdpi.com/1999-5903/17/4/138/pdf
    Download Restriction: no

    File URL: https://www.mdpi.com/1999-5903/17/4/138/
    Download Restriction: no
    ---><---

    References listed on IDEAS

    as
    1. Karan Singhal & Shekoofeh Azizi & Tao Tu & S. Sara Mahdavi & Jason Wei & Hyung Won Chung & Nathan Scales & Ajay Tanwani & Heather Cole-Lewis & Stephen Pfohl & Perry Payne & Martin Seneviratne & Paul G, 2023. "Publisher Correction: Large language models encode clinical knowledge," Nature, Nature, vol. 620(7973), pages 19-19, August.
    2. Cinelli, Marco & Kadziński, Miłosz & Gonzalez, Michael & Słowiński, Roman, 2020. "How to support the application of multiple criteria decision analysis? Let us start with a comprehensive taxonomy," Omega, Elsevier, vol. 96(C).
    3. Kuo, Ting, 2017. "A modified TOPSIS with a different ranking index," European Journal of Operational Research, Elsevier, vol. 260(1), pages 152-160.
    4. Karan Singhal & Shekoofeh Azizi & Tao Tu & S. Sara Mahdavi & Jason Wei & Hyung Won Chung & Nathan Scales & Ajay Tanwani & Heather Cole-Lewis & Stephen Pfohl & Perry Payne & Martin Seneviratne & Paul G, 2023. "Large language models encode clinical knowledge," Nature, Nature, vol. 620(7972), pages 172-180, August.
    Full references (including those not matched with items on IDEAS)

    Most related items

    These are the items that most often cite the same works as this one and are cited by the same works as this one.
    1. Maxime Griot & Coralie Hemptinne & Jean Vanderdonckt & Demet Yuksel, 2025. "Large Language Models lack essential metacognition for reliable medical reasoning," Nature Communications, Nature, vol. 16(1), pages 1-10, December.
    2. Ching-Nam Hang & Pei-Duo Yu & Roberto Morabito & Chee-Wei Tan, 2024. "Large Language Models Meet Next-Generation Networking Technologies: A Review," Future Internet, MDPI, vol. 16(10), pages 1-29, October.
    3. Francesco Ciardiello & Andrea Genovese, 2023. "A comparison between TOPSIS and SAW methods," Annals of Operations Research, Springer, vol. 325(2), pages 967-994, June.
    4. Susmaga, Robert & Szczȩch, Izabela & Zielniewicz, Piotr & Brzezinski, Dariusz, 2023. "MSD-space: Visualizing the inner-workings of TOPSIS aggregations," European Journal of Operational Research, Elsevier, vol. 308(1), pages 229-242.
    5. Tingmingke Lu, 2025. "Maximum Hallucination Standards for Domain-Specific Large Language Models," Papers 2503.05481, arXiv.org.
    6. Zheng, Shuwen & Pan, Kai & Liu, Jie & Chen, Yunxia, 2024. "Empirical study on fine-tuning pre-trained large language models for fault diagnosis of complex systems," Reliability Engineering and System Safety, Elsevier, vol. 252(C).
    7. van Kolfschooten, Hannah & van Oirschot, Janneke, 2024. "The EU Artificial Intelligence Act (2024): Implications for healthcare," Health Policy, Elsevier, vol. 149(C).
    8. Soroosh Tayebi Arasteh & Tianyu Han & Mahshad Lotfinia & Christiane Kuhl & Jakob Nikolas Kather & Daniel Truhn & Sven Nebelung, 2024. "Large language models streamline automated machine learning for clinical studies," Nature Communications, Nature, vol. 15(1), pages 1-12, December.
    9. Hossam A. Gabber & Omar S. Hemied, 2024. "Domain-Specific Large Language Model for Renewable Energy and Hydrogen Deployment Strategies," Energies, MDPI, vol. 17(23), pages 1-25, December.
    10. Zhenjia Chen & Zhenyuan Lin & Ji Yang & Cong Chen & Di Liu & Liuting Shan & Yuanyuan Hu & Tailiang Guo & Huipeng Chen, 2024. "Cross-layer transmission realized by light-emitting memristor for constructing ultra-deep neural network with transfer learning ability," Nature Communications, Nature, vol. 15(1), pages 1-12, December.
    11. Yujin Oh & Sangjoon Park & Hwa Kyung Byun & Yeona Cho & Ik Jae Lee & Jin Sung Kim & Jong Chul Ye, 2024. "LLM-driven multimodal target volume contouring in radiation oncology," Nature Communications, Nature, vol. 15(1), pages 1-14, December.
    12. Chen Gao & Xiaochong Lan & Nian Li & Yuan Yuan & Jingtao Ding & Zhilun Zhou & Fengli Xu & Yong Li, 2024. "Large language models empowered agent-based modeling and simulation: a survey and perspectives," Palgrave Communications, Palgrave Macmillan, vol. 11(1), pages 1-24, December.
    13. Juexiao Zhou & Xiaonan He & Liyuan Sun & Jiannan Xu & Xiuying Chen & Yuetan Chu & Longxi Zhou & Xingyu Liao & Bin Zhang & Shawn Afvari & Xin Gao, 2024. "Pre-trained multimodal large language model enhances dermatological diagnosis using SkinGPT-4," Nature Communications, Nature, vol. 15(1), pages 1-12, December.
    14. Khannoussi, Arwa & Meyer, Patrick & Chaubet, Aurore, 2023. "A multi-criteria decision aiding approach for upgrading public sewerage systems and its application to the city of Brest," Socio-Economic Planning Sciences, Elsevier, vol. 87(PA).
    15. Junyi Chai & Zhiquan Weng & Wenbin Liu, 2021. "Behavioral Decision Making in Normative and Descriptive Views: A Critical Review of Literature," JRFM, MDPI, vol. 14(10), pages 1-14, October.
    16. Zhang, Sheng & Lin, Zhang & Ai, Zhengtao & Huan, Chao & Cheng, Yong & Wang, Fenghao, 2019. "Multi-criteria performance optimization for operation of stratum ventilation under heating mode," Applied Energy, Elsevier, vol. 239(C), pages 969-980.
    17. Pätäri, Eero & Karell, Ville & Luukka, Pasi & Yeomans, Julian S, 2018. "Comparison of the multicriteria decision-making methods for equity portfolio selection: The U.S. evidence," European Journal of Operational Research, Elsevier, vol. 265(2), pages 655-672.
    18. Xiaodong Yuan & Weiling Song, 2022. "Evaluating technology innovation capabilities of companies based on entropy- TOPSIS: the case of solar cell companies," Information Technology and Management, Springer, vol. 23(2), pages 65-76, June.
    19. Ricardo, Alexandre & Figueira, José Rui & Tavares, Luís Valadares, 2024. "Integrating confidence and preservation of information in the preference elicitation process: A lexicographic order approach for inconsistent judgments," Omega, Elsevier, vol. 129(C).
    20. Wang, Xiaoli & Zhang, Chenxi & Xu, Zeshui, 2024. "A product recommendation model based on online reviews: Improving PageRank algorithm considering attribute weights," Journal of Retailing and Consumer Services, Elsevier, vol. 81(C).

    Corrections

    All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:gam:jftint:v:17:y:2025:i:4:p:138-:d:1618786. See general information about how to correct material in RePEc.

    If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.

    If CitEc recognized a bibliographic reference but did not link an item in RePEc to it, you can help with this form .

    If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.

    For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: MDPI Indexing Manager (email available below). General contact details of provider: https://www.mdpi.com .

    Please note that corrections may take a couple of weeks to filter through the various RePEc services.

    IDEAS is a RePEc service. RePEc uses bibliographic data supplied by the respective publishers.