融合动态标签平滑策略和关系网络的中医证候分类模型

doi:10.3778/j.issn.1002-8331.2407-0408

摘要/Abstract

摘要： 中医证候智能分类研究中，由于类别不平衡和高质量人工标注样本较少，模型在学习少样本标签时能力不佳，导致整体分类效果不理想。为解决以上问题，提出了一种融合动态标签平滑策略和关系网络的中医证候分类模型（online label smoothing for relational networks based on pre-trained language models，PLM-SNet）。该模型使用预训练语言模型对输入的病例文本进行编码，获取输入样本的特征表示；通过关系网络对样本的支持集和查询集进行特征信息的级联，获得查询集样本的相关性得分；采用动态标签监督模型的训练损失，实时更新和优化类别的软标签，从而获得最终的类别得分。在中医公开数据集TCM-SD和自建中医哮喘数据集J-SD上的实验结果表明，与最优基线模型相比，PLM-SNet将中医证候分类的Macro-F1和G-Mean分别提高了3.47、2.48、3.06和2.58个百分点，实验结果验证了该模型在类别不平衡中医证候分类任务中的科学性和有效性。

关键词: 动态标签, 关系网络, 类别不平衡, 中医证候分类

Abstract: In the research on intelligent classification of TCM syndromes, due to class imbalance and the scarcity of high-quality manually labeled samples, the ability of model to learn from few-sample labels is inadequate, resulting in unsatisfactory overall classification performance. To address these issues, an online label smoothing strategy integrated with relational networks for TCM syndrome classification models (online label smoothing for relational networks based on pre-trained language models, PLM-SNet) is proposed. This model encodes the input case text using pre-trained language models to obtain feature representations of the input samples. It cascades the feature information through a relational network using the support set and query set of the samples to obtain the relevance scores of the samples in the query set. The training loss of the online label supervision model is used to update and optimize the soft labels of the categories in real-time, resulting in the final category scores. Experimental results on the TCM public dataset TCM-SD and the self-constructed TCM asthma dataset J-SD show that compared to the optimal baseline model, PLM-SNet improves the Macro-F1 and G-Mean of TCM syndrome classification by 3.47, 2.48, 3.06, and 2.58 percentage points, respectively. The experimental results verify the scientific validity and effectiveness of the model in the class imbalance TCM syndrome classification task.

Key words: online label, relational networks, category imbalance, Chinese medicine syndrome classification

廖明, 杜建强, 罗计根, 黄强, 贺佳, 范越. 融合动态标签平滑策略和关系网络的中医证候分类模型[J]. 计算机工程与应用, 2025, 61(21): 351-360.

LIAO Ming, DU Jianqiang, LUO Jigen, HUANG Qiang, HE Jia, FAN Yue. Fusion of Online Label Smoothing Strategy and Relational Network for TCM Syndrome Classification Model[J]. Computer Engineering and Applications, 2025, 61(21): 351-360.

参考文献

[1] LI W Y, GE X L, LIU S, et al. Opportunities and challenges of traditional Chinese medicine doctors in the era of artificial intelligence[J]. Frontiers in Medicine, 2024, 10: 1336175.
[2] AMIRI Z, HEIDARI A, DARBANDI M, et al. The personal health applications of machine learning techniques in the Internet of behaviors[J]. Sustainability, 2023, 15(16): 12406.
[3] 曾江楠, 叶廷林, 李敏芳, 等. 新型冠状病毒感染后咳嗽中医证型和体质分布规律[J]. 辽宁中医杂志, 2025, 52(2): 1-4.
ZENG J N, YE T L, LI M F, et al. TCM syndrome types and constitution distribution of cough after CoronaVirus disease 2019[J]. Liaoning Journal of Traditional Chinese Medicine, 2025, 52(2): 1-4.
[4] CHEN Z, ZHANG D, LIU C X, et al. Traditional Chinese medicine diagnostic prediction model for holistic syndrome differentiation based on deep learning[J]. Integrative Medicine Research, 2024, 13(1): 101019.
[5] LI X C, CHEN K, YANG J X, et al. TLDA: a transfer learning based dual-augmentation strategy for traditional Chinese medicine syndrome differentiation in rare disease[J]. Computers in Biology and Medicine, 2024, 169: 107808.
[6] REN M C, HUANG H Y, ZHOU Y X, et al. TCM-SD: a benchmark for probing syndrome differentiation via natural language processing[C]//Proceedings of the 21st Chinese National Conference on Computational Linguistics Chinese. Cham: Springer International Publishing, 2022: 247-263.
[7] RICHTER T, NESTLER-PARR S, BABELA R, et al. Rare disease terminology and definitions: a systematic global review: report of the ISPOR rare disease special interest group[J]. Value in Health, 2015, 18(6): 906-914.
[8] ZHANG C B, JIANG P T, HOU Q B, et al. Delving deep into label smoothing[J]. IEEE Transactions on Image Processing, 2021, 30: 5984-5996.
[9] HU C Y, ZHANG S Y, GU T Y, et al. Multi-task joint learning model for Chinese word segmentation and syndrome differentiation in traditional Chinese medicine[J]. International Journal of Environmental Research and Public Health, 2022, 19(9): 5601.
[10] ZHANG H, ZHANG J J, NI W D, et al. Transformer-and generative adversarial network-based inpatient traditional Chinese medicine prescription recommendation: development study[J]. JMIR Medical Informatics, 2022, 10(5): 35239.
[11] LIU Z Q, HE H Y, YAN S X, et al. End-to-end models to imitate traditional Chinese medicine syndrome differentiation in lung cancer diagnosis: model development and validation[J]. JMIR Medical Informatics, 2020, 8(6): 17821.
[12] MIN B N, ROSS H, SULEM E, et al. Recent advances in natural language processing via large pre-trained language models: a survey[J]. ACM Computing Surveys, 2024, 56(2): 1-40.
[13] WANG H F, LI J W, WU H, et al. Pre-trained language models and their applications[J]. Engineering, 2023, 25: 51-65.
[14] YAO L, JIN Z, MAO C S, et al. Traditional Chinese medicine clinical records classification with BERT and domain specific corpora[J]. Journal of the American Medical Informatics Association, 2019, 26(12): 1632-1636.
[15] ZHANG X, CHEN Z K, GAO J, et al. A two-stage deep transfer learning model and its application for medical image processing in traditional Chinese medicine[J]. Knowledge-Based Systems, 2022, 239: 108060.
[16] LI J Y, TANG T Y, ZHAO W X, et al. Pre-trained language models for text generation: a survey[J]. ACM Computing Surveys, 2024, 56(9): 1-39.
[17] DEVLIN J, CHANG M W, LEE K, et al. BERT: pre-training of deep bidirectional transformers for language understanding[C]//Proceedings of the North American Chapter of the Association for Computational Linguistics, 2019.
[18] SUNG F, YANG Y X, ZHANG L, et al. Learning to compare: relation network for few-shot learning[C]//Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2018: 1199-1208.
[19] 苏翀, 任曈, 王国品, 等. 利用决策树建立慢性阻塞性肺病中医诊断模型[J]. 计算机工程与应用, 2019, 55(3): 225-230.
SU C, REN T, WANG G P, et al. Using K-L divergence based decision tree to build traditional Chinese medicine diagnosis model on COPD[J]. Computer Engineering and Applications, 2019, 55(3): 225-230.
[20] LAN Z Z, CHEN M D, GOODMAN S, et al. ALBERT: a lite BERT for self-supervised learning of language representations[J]. arXiv:1909.11942, 2019.
[21] LIU Y H, OTT M, GOYAL N, et al. RoBERTa: a robustly optimized BERT pretraining approach[J]. arXiv:1907.11692, 2019.