Transformer-Based Classification of User Queries for Medical Consultancy

Cover Page

Cite item

Full Text

Open Access Open Access
Restricted Access Access granted
Restricted Access Subscription Access

Abstract

Представлен новый подход, использующий модель RuBERT для классификации пользовательских запросов в области медицинских консультаций с учетом специализации эксперта. В ходе исследования был собран обширный набор данных, который использовался для дообучения модели RuBERT. Метрика качества полученной модели F1-score составила более 91,8% как при использовании блоковой кросс-валидации, так и при разделении набора данных на обучающую и тестовую выборки. Подход демонстрирует высокую обобщающую способность для различных медицинских подобластей, таких как кардиология, неврология и дерматология. Предложенный подход позволяет сократить время на определение наиболее подходящего специалиста и тем самым повышает качество консультации и медицинской помощи.

References

  1. Trusting Social Media as a Source of Health Information: Online Surveys Comparing the United States, Korea, and Hong Kong / H. Song // J. Medic. Internet Res. 2016. V. 18. No. 3. P. 25. URL: https://www.jmir.org/2016/3/e25. https://doi.org/10.2196/jmir.4193
  2. БэбиБлог — Ответы на любые вопросы о беременности, детях и семейной жизни. Accessed: December 19 , 2022. https://www.babyblog.ru/
  3. Keshavarz H. Evaluating credibility of social media information: current challenges, research directions and practical criteria // Inform. Discover. Deliver. 2021. V. 49. No. 4. P. 269–279. https://doi.org/10.1108/IDD-03-2020-0033
  4. Automatic medical specialty classification based on patients’ description of their symptoms / C. Mao / BMC Medical Informatics and Decision Making. 2023. V. 23. https://doi.org/10.1186/s12911-023-02105-7
  5. Tezgider M., Yildiz B., Aydin G. Text classification using improved bidirectional transformer // Concurrency and Computation: Practice and Experience. 2022. V. 34. No. 9. eprint: https://onlinelibrary.wiley.com/doi/pdf/10.1002/cpe.6486. URL: https://onlinelibrary.wiley.com/doi/abs/10.1002/cpe.6486. https://doi.org/https://doi.org/10.1002/cpe.6486
  6. СпросиВрача: Задай вопрос врачу онлайн и получи ответ мгновенно. Accessed: February 17, 2023. https://sprosivracha.com/
  7. ДОКТУ — поиск лучших врачей и клиник в России. Accessed: February 17, 2023. https://doctu.ru/
  8. Онлайн — медицинские консультации в режиме онлайн. Accessed: February 17, 2023. https://03online.com/
  9. health.mail.ru — Поиск по болезням, лекарствам и ответам врачей. Accessed: February 17, 2023. https://health.mail.ru/
  10. Johnson J.M., Khoshgoftaar T.M. Survey on deep learning with class imbalance // Journal of Big Data. 2019. V. 6. No. 1. P. 27. https://doi.org/10.1186/s40537-019-0192-5
  11. Ma E. NLP Augmentation. 2019. Accessed: February 17, 2023. https://github.com/makcedward/nlpaug
  12. Hecht-Nielsen R. III.3 – Theory of the Backpropagation Neural Network (Based on “nonindent” by Robert Hecht-Nielsen, which appeared in Proceedings of the International Joint Conference on Neural Networks 1, 593–611, June 1989). ×c 1989 IEEE / Neural Networks for Perception / H. Wechsler (Ed.). Academic Press, 1992. P. 65–93. ISBN 978-0-12-741252-8. https://doi.org/10.1016/B978-0-12-741252-8.50010-8. URL: https://www.sciencedirect.com/science/article/pii/B9780127412528500108
  13. Shaheen Z., Wohlgenannt G., Filtz E. Large Scale Legal Text Classification Using Transformer Models. 2020. arXiv: 2010.12871 [cs.CL]
  14. Understanding AdamW through Proximal Methods and Scale-Freeness / Z. Zhuang. 2022. arXiv: 2202.00089 [cs.LG]
  15. Automated Learning Rate Scheduler for Large-batch Training / C. Kim. 2021. arXiv: 2107.05855 [cs.LG]
  16. Attention Is All You Need / A. Vaswani. 2017. arXiv: 1706.03762 [cs.CL]
  17. Large Batch Optimization for Deep Learning: Training BERT in 76 minutes / Y. You. 2020. arXiv: 1904.00962 [cs.LG]
  18. Are Transformers more robust than CNNs? / Y. Bai // Advances in Neural Information Processing Systems. 2021. P. 34. Curran Associates, Inc. P. 26831–26843. URL: https://proceedings.neurips.cc/paper files/paper/2021/file/ e19347e1c3ca0c0b97de5fb3b690855a
  19. A Survey on Text Classification: From Shallow to Deep Learning / Q. Li. 2021. arXiv: 2008.00364 [cs.CL]
  20. Transformers: State-of-the-Art Natural Language Processing / T. Wolf [et al.] // Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing: System Demonstrations. Online : Association for Computational Linguistics. 2020. P. 38–45. URL: https://www.aclweb.org/anthology/2020.emnlp-demos.6
  21. Maida A. Cognitive Computing and Neural Networks: Reverse Engineering the Brain / Handbook of Statistics. V. 35. Elsevier. 2016. P. 39–78. https://doi.org/10.1016/bs.host.2016.07.011 URL: https://doi.org/10.1016/bs.host.2016.07.011
  22. Kostenetskiy P.S., Chulkevich R.A., Kozyrev V.I. HPC Resources of the Higher School of Economics / J. Physics: Conf. 2021. P. 1740. No. 1. P. 012050. https://doi.org/10.1088/1742-6596/1740/1/012050 URL: https://dx.doi.org/10.1088/1742-6596/1740/1/012050
  23. Reimers N., Gurevych I. Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks. 2019. arXiv: 1908.10084 [cs.CL]
  24. Language-agnostic BERT Sentence Embedding / F. Feng. 2022. arXiv: 2007.01852 [cs.CL]
  25. Kuratov Y., Arkhipov M. Adaptation of Deep Bidirectional Multilingual Transformers for Russian Language. 2019. arXiv: 1905.07213 [cs.CL]
  26. BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding / J. Devlin. 2019. arXiv: 1810.04805 [cs.CL]
  27. BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension / M. Lewis. 2019. arXiv: 1910.13461 [cs.CL]

Copyright (c) 2024 The Russian Academy of Sciences

This website uses cookies

You consent to our cookies if you continue to use our website.

About Cookies