Article Details

Title · Authors · Abstract · Keywords · DOI · References

Application and Challenges of Deep Learning Technology in Text Mining of Traditional Chinese Medicine Formulas

LI Yiwen, LI Zhibiao, LIU Zixuan

(School of Intelligent Medicine and Information Engineering, Jiangxi University of Chinese Medicine, Nanchang 330004, China)

Section：Computer Technology · Vol. 2026, Issue 4 · Pages：73-80 · 2096-4706（2026）04-0073-08 · CLC number：TP18 · Document code：A · DOI：10.19850/j.cnki.2096-4706.2026.04.013

Abstract

Traditional Chinese Medicine (TCM) formulas have developed over thousands of years and embody the precious experience and intellectual crystallization of medical practitioners across all dynasties. In recent years, the rapid development of Deep Learning technology has provided an effective approach for exploring their inherent logic. This paper sorts out and analyzes the applications of Deep Learning in TCM formula research and classifies it by technical category. Faced with the inherently complex knowledge system of TCM formulas, Deep Learning technology exhibits its unique advantages, especially in efficacy classification, formula composition analysis and novel drug discovery. However, current research still faces challenges such as limited and unbalanced data, complex feature extraction, great difficulty in model construction and insufficient integration of TCM knowledge. In the future, research can advance around the construction of standardized formula datasets, multimodal data fusion, the driving of cutting-edge Artificial Intelligence technologies and interdisciplinary integration, so as to further explore the mechanism of action and scientific connotation of TCM formulas.

Keywords

Traditional Chinese Medicine formula; text mining; Deep Learning; Large Language Model; Artificial Intelligence

References

  1. [1] Zi Lei, Li Yanjuan, Li Yanqin, etc Progress in research methods/strategies for the compatibility mechanism of traditional Chinese medicine formulas [J]. Chinese Pharmacy, 2023, 34 (11): 1393-1398
  2. [2] An Li, Tan Feng, Li Mengwen, etc Exploring Several Issues in the Study of Prescription Based on Disciplinary Characteristics [J]. Chinese Journal of Traditional Chinese Medicine, 2024, 39 (2): 609-612
  3. [3] BAO H Y，WEN R J，LI X Y，et al. A Brief Overview of Traditional Chinese Medicine Prescription Powered by Artificial Intelligence [J/OL].TMR Modern Herbal Medicine，2021，4（2）：13（2021-04-27）.https://doi.org/10.53388/MHM2021B0412001.
  4. [4] Gao Wanqing, Cheng Ning, Li Lisong, etc Overview of the Application of Machine Learning in Formula Research [J]. Chinese Journal of Traditional Chinese Medicine Information, 2021, 28 (6): 131-137
  5. [5] HE X，SUN M，BATTULGA T，et al. Application of Artificial Intelligence in the Development of Traditional Chinese Medicine [J/OL].Basic & Clinical Pharmacology & Toxicology，2025，137（1）：e70066（2025-06-09）.https://doi.org/10.1111/bcpt.70066.
  6. [6] Geng Jiahao Research and Application of Traditional Chinese Medicine Intelligent Prescription Based on Deep Learning [D]. Harbin: Heilongjiang University, 2024
  7. [7] SHARIFANI K，AMINI M. Machine Learning and Deep Learning: A Review of Methods and Applications [J].World Information Technology and Engineering Journal，2023，10（7）：3897-3904.
  8. [8] JANIESCH C，ZSCHECH P，HEINRICH K. Machine Learning and Deep Learning [J].Electronic Markets，2021，31（3）：1-11.
  9. [9] LECUN Y，BENGIO Y，HINTON G. Deep learning [J].Nature，2015，521（7553）：436-444.
  10. [10] Liu Xing, Shen Lu The dose effect relationship and compatibility rules of dried ginger in "Treatise on Cold Damage and Miscellaneous Diseases" [J]. World Journal of Traditional Chinese Medicine, 2023, 18 (6): 865-870876
  11. [11] Zhou Ying, Zhou Yang Analysis of the characteristics of gynecological disease medication in the "Shennong Bencao Jing" [J]. Chinese Journal of Traditional Chinese Medicine, 2024, 39 (2): 1016-1021
  12. [12] Kong Xiaoli, Peng Mengfan, Zhou Xinxin, et al. Analysis and thinking on the characteristics of traditional Chinese patent medicines and simple preparations for regulating menstruation in the Chinese Pharmacopoeia (2020 edition) [J]. Chinese Journal of Experimental Prescriptions, 2023, 29 (14): 173-180
  13. [13] ZHANG P，SHEN S W，DENG W P，et al. The Construction Model of the TCM Clinical Knowledge Coding Database Based on Knowledge Organization [J/OL].BioMed research international，2022，2022：2503779（2022-01-17）.https://doi.org/10.1155/2022/2503779.
  14. [14] REN X，WANG H，DAI D. A Summary of Research on Web Data Acquisition Methods based on Distributed Crawler [C]//2020 IEEE 6th International Conference on Computer and Communications.Chengdu：IEEE，2020：1682-1688.
  15. [15] MUSTAFA T M. Theoretical Understanding of Convolutional Neural Network: Concepts，Architectures，Applications，Future Directions [J/OL].Computation，2023，11（3）：52（2023-03-06）.https://doi.org/10.3390/computation11030052
  16. [16] ALZUBAIDI L，ZHANG J L，HUMAIDI A J，et al. Review of Deep Learning: Concepts, CNN Architectures, Challenges, Applications, Future Diretions [J/OL].Journal of Big Data，2021，8（1）：53（2021-03-31）.https://doi.org/10.1186/s40537-021-00444-8.
  17. [17] ZHOU W，YANG K，ZENG J Y，et al. FordNet: Recommending Traditional Chinese Medicine Formula Via Deep Neural Network Integrating Phenotype and Molecule [J/OL].Pharmacol Res，2021，173：105752（2021-09-09）.https://doi.org/10.1016/j.phrs.2021.105752.
  18. [18] Chen Canyu, Fu Amin, Li Guowei, etc Chinese Herbal Medicine Recommendation System Based on Convolutional Neural Network [J]. World Science and Technology - Modernization of Traditional Chinese Medicine, 2022, 24 (6): 2402-2410
  19. [19] LI Z B，ZHAO H Y，ZHU G H，et al. Classification Method of Traditional Chinese Medicine Compound Decoction Duration based on Multi-Dimensional Feature Weighted Fusion [J].Computer Methods in Biomechanics and Biomedical Engi-neering，2025，28（6）：867-881.
  20. [20] Yang Xiaohuan Research on Extracting the Relationship between Traditional Chinese Medicine Formulas and Diseases for Abstracts [D]. Wuhan: Hubei University of Traditional Chinese Medicine, 2018
  21. [21] LI W，YANG Z. Exploration on Generating Traditional Chinese Medicine Prescriptions from Symptoms with an End-to-End Approach [C]//Proceedings of the 8th CCF International Conference on Natural Language Processing and Chinese Computing.Dunhuang：Springer International Publishing，2019：486-498.
  22. [22] WAQAS M，HUMPHRIES U W.A Critical Review of RNN and LSTM Variants in Hydrological Time Series Predictions [J/OL].MethodsX，2024，13：102946（2024-09-17）.https://doi.org/10.1016/j.mex.2024.102946.
  23. [23] GOODFELLOW I，POUGET A J，MIRZA M，et al. Generative Adversarial Networks [J].Communications of the ACM，2020，63（11）：139-144.
  24. [24] RONG C，LI X，SUN X，et al. Chinese Medicine Prescription Recommendation using Generative Adversarial Network [J].IEEE Access，2022，10：12219-12228.
  25. [25] Hao Yifeng Named Entity Recognition for Traditional Chinese Medicine Text Based on GAN [D]. Tangshan: North China University of Science and Technology, 2022
  26. [26] AFIA S，MAINUL H B. Exploring Generative Adversarial Networks and Adversarial Training [J].International Journal of Cognitive Computing in Engineering，2022（prepublish）：78-89.
  27. [27] ASIF N A，SARKER Y，CHAKRABORTTY R K，et al. Graph Neural Network: A Comprehensive Review on Non-Euclidean Space [J].IEEE ACCESS，2021，9：60588-60606.
  28. [28] Liu Jiajun Analysis of prescription efficacy based on graph neural network [D]. Changsha: Hunan University of Traditional Chinese Medicine, 2023
  29. [29] Zhao Danning, Lu Haixia, Bai Sen Classification and automatic recommendation of traditional Chinese medicine prescription efficacy based on graph convolutional neural network [J]. Chinese Herbal Medicine, 2024, 55 (18): 6298-6304
  30. [30] LI W，YANG Z，SHAO Y. Domain Specific Pre-training Methods for Traditional Chinese Medicine Prescription Recommendation [C]//CAAI Inter-national Conference on Artificial Intelligence.Singapore：Springer Nature Singapore，2023：125-135.
  31. [31] YU F，TANG J，YIN W，et al. Ernievil: Knowledge Enhanced Visionlanguage Representations Through Scene Graphs [C]//Proceedings of the AAAI Conference on Artificial Intelligence. 2021：3208-3216.
  32. [32] XUE Q，GAO B，WEN J，et al. A Deep Learning-Based Herb Pair Discovering Approach [C]//CCF Conference on Big Data.Singapore：Springer Nature Singapore，2021：135-149.
  33. Guo Fengying, Gao Chenxi, Wang Yifan, etc Research on Intelligent Recommendation Method of Ancient Prescriptions Based on Deep Learning and Medical Case Similarity [J]. Chinese Journal of Traditional Chinese Medicine, 2025, 43 (6): 24-30
  34. [34] HASSAN M M，GUMAEI A，ALSANAD A，et al. A Hybrid Deep Learning Model for Efficient Intrusion Detection in Big Data Environment [J].Information Sciences，2020，513：386-396.
  35. [35] JING N，WU Z，WANG H F.A Hybrid Model Integrating Deep Learning with Investor Sentiment Analysis for Stock Price Prediction [J].Expert Systems With Applications，2021，178：115019.
  36. [36] SALUR M U，AYDIN I.A Novel Hybrid Deep Learning Model for Sentiment Classification [J].IEEE Access，2020，8：58080-58093.
  37. [37] TIAN B，GUAN H T，SHANG W，et al. Traditional Chinese Medicine Entity Relation Extraction based on CNN with Segment Attention [J].Neural Computing and Applications，2021（prepublish）：1-10.
  38. [38] CHENG N，CHEN Y，GAO W Q，et al. An Improved Deep Learning Model: S-TextBLCNN for Traditional Chinese Medicine Formula Classification [J/OL].Frontiers in Genetics，2021，12：807825（2021-12-22）.https://doi.org/10.3389/fgene.2021.807825.
  39. [39] QU Q Q，KAN H X，WU Y T，et al. Named Entity Recognition of TCM Text Based on Bert Model [C]//2020 7th International Forum on Electrical Engineering and Automation (IFEEA).Hefei：IEEE，2020：652-655.
  40. [40] REN Q Y，LI K Q，YANG D S，et al. TCM Function Multi-classification Approach Using Deep Learning Models [C]//International Conference on Web Information Systems and Applications.Singapore：Springer Nature Singapore，2023：246-258.
  41. [41] ZOU S L，LIU Z Y，WANG K Q，et al. A Study on Pharmaceutical Text Relationship Extraction based on Heterogeneous Graph Neural Networks [J].Math Biosci Eng，2024，21（1）：1489-1507.
  42. [42] SEDIR M，BUDACH L，FEUERPFEIL M，et al. The Effects of Data Quality on Machine Learning Performance on Tabular Data [J/OL].Information Systems，2025，132：102549（2025-03-18）.https://doi.org/10.1016/j.is.2025.102549.
  43. [43] Sun Yanjie Research on Data Enhancement Algorithm for Unbalanced Data Classification Problem [D]. Changchun: Jilin University, 2023
  44. [44] Li Ang, Han Meng, Mu Dongliang, etc A review of classification methods for multi class imbalanced data [J]. Computer Application Research, 2022, 39 (12): 3534-3545
  45. Xiao Wenke, Song Chi, Chen Shilin, etc Key Technologies and Construction Strategies of Traditional Chinese Medicine Language Model [J]. Chinese Herbal Medicine, 2024, 55 (17): 5747-5756
  46. [46] ZHANG T T，HUANG Z H，WANG Y Q，et al. Information Extraction from the Text Data on Traditional Chinese Medicine: A Review on Tasks, Challenges, and Methods from 2010 to 2021 [J/OL]. Evidence-Based Complementary and Al-ternative Medicine，2022（1）：1679589（2022-05-13）.https://doi.org/10.1155/2022/1679589.
  47. [47] Tang Heng'an, Tang Sange, Tang Zhishu On the Second Abstraction of Traditional Chinese Medicine Thought System [J]. Chinese Journal of Traditional Chinese Medicine, 2022, 37 (9): 4897-4902
  48. Shen Wang, Liu Jiayu, Li He, etc Research progress on digitalization of traditional Chinese medicine ancient books from the perspective of disciplinary life cycle in China [J]. Library and Information Work, 2022, 66 (22): 4-15
  49. [49] ZHANG C Y，BENGIO S，HARDT M，et al. Understanding Deep Learning (Still) Requires Rethinking Generalization [J].COMMUNICATIONS OF THE ACM，2021，64（3）：107-115.
  50. [50] ZHANG Y Q.A Comprehensive Survey on Regularization Strategies in Machine Learning [J].Information Fusion，2022，80：146-166.
  51. [51] MORADI R，BERANGI R，MINAEI B. A Survey of Regularization Strategies for Deep Models [J].Artificial Intelligence Rev，2020，53（6）：3947-3986.
  52. [52] BANSAL A，SHARMA R，KATHURIA M . A Systematic Review on Data Scarcity Problem in Deep Learning：Solution and Applications [J].ACM Computing Surveys (CSUR)，2022，54（10）：1-29.
  53. [53] ZINI J E，AWAD M. On the Explainability of Natural Language Processing Deep Models [J].ACM Computing Surveys，2022，55（5）：1-31.
  54. [54] WU Z Y，PENG S Y，ZHU Y，et al. Formula-S: Situated Visualization for Traditional Chinese Medicine Formula Learning [J].Chinese Medical Sciences Journal，2025，40（1）：57-67.
  55. [55] Wang Guangyao, Yang Haixia, Wang Xinghua, etc The Approaches and Methods of Basic Theory Research in Traditional Chinese Medicine [J]. Chinese Journal of Traditional Chinese Medicine, 2023, 38 (2): 691-694

Funding

Key Discipline Construction Fund of Jiangxi University of Traditional Chinese Medicine (2023jzzdxk025)

About the authors

Li Yiwen (2001-), female, Han ethnicity, from Xinyu, Jiangxi, currently pursuing a master's degree in natural language processing; Corresponding author: Li Zhibiao (1981-), male, Han ethnicity, from Fuzhou, Jiangxi Province, associate professor, doctoral student, research direction: Traditional Chinese Medicine Data Mining.