复合结构智能化辨证选方模型的构建  

Construction of Intelligent Syndrome Differentiation and Formula Selection of Compound Structure Model

作　　者：周璐 李光庚 孙燕 郑岩[2] 李宇航 Zhou Lu;Li Guanggeng;Sun Yan;Zheng Yan;Li Yuhang(School of Chinese Medicine, Beifing University of Chinese Medicine, Belling 100029, China;School of Computer Science, Beijing University of Posts and Telecommunications, Beijing 100876, China)

机构地区：[1]北京中医药大学中医学院,北京100029 [2]北京邮电大学计算机学院,北京100876

出　　处：《世界中医药》2018年第2期479-483,共5页World Chinese Medicine

基　　金：国家重点研发计划"中医药现代化研究"课题(2017YFC1700303)

摘　　要：通过回顾机器学习中的C4.5决策树、随机森林、支持向量机、BP神经网络算法的原理与在中医辨证研究中所取得的成果,创新设计出一种复合结构的智能化辨证选方模型,并对模型进行了实现与测试。结果表明该模型输出结果的准确性高于仅使用单一算法建立的辨证选方模型,这为进一步与"方-证要素对应"的组方原则相结合,建立适用于复杂病机的临床诊疗辅助系统奠定了基础。Through a review of the C4. 5 Decision Tree,Random Forest,Support Vector Machine and BP Neural Network algorithm and the obtained in the study of TCM syndrome differentiation,a compound structure of intelligent syndrome differentiation and formula selection model was designed. The model was implemented and tested. Results showed that the accuracy of the corresponding results was higher than that of the single algorithm,laying the foundation for the model which was suitable for the complex Pathogenesis of clinical diagnosis and treatment of auxiliary system.

关 键 词：人工智能 机器学习 方-证要素对应 辨证论治 

分 类 号：R241[医药卫生—中医诊断学] R242[医药卫生—中医临床基础]