缺失药性补遗与完备的方法学研究

Methodological Study of the Unrecorded Nature of Chinese Medicine and Associated Remedy

目的:在中医药理论的指导下,运用数据挖掘技术研究中药功能与药性的关系,建立基于功能的药性预测模型,通过模型预测药物缺失的药性,为药性研究与完备提供支持。方法:以2000版《中华人民共和国药典》(一部)中功能、药性数据为依据,利用决策树学习算法。分别对功能与四气、功能与五味、功能与归经间的关系进行研究,建立预测模型;根据相应模型,对缺失药性进行预测。结果:建立四气、五味、归经预测模型共20个,表明多数模型具有较好的可信度;利用模型对药性的预测结果与文献报道基本一致。结论:所建立预测模型可用于缺失药性的补遗与完备。

This paper studies the relationship between the Functions of Chinese Medicine （FCM） and the Nature of Chinese Medicine （NCM）. The study is conducted under the guidance of the Theory of Traditional Chinese Medicine（TTCM）, with the help of the data mining technology. A NCM performance Forecasting Model （FM） based on FCM is established to improve the understanding of NCM and associated study. Authors searched the database of 《PHARMACOPOEIA OF THE PEOPLE＇S REPUBLIC OF CHINA 》（2000 Edition）, and found 507 kinds of Chinese medicines enjoying the records of both FCM and NCM. With the help of data mining technology, the relationships between FCM and Siqi, FCM and Wuwei, and FCM and Guijing are examined to establish the valid FM. More NCM information can be sorted out, with the help of FM. As a result, 20 models have been established, including siqi, wuwei, and guijing. Some of them have been proved reliable according to the results. Obviously, there are some regular links between FCM and NCM. With the help of the FM, this paper predicts some unknown information of 19 Chinese Medicines, including Herba Hyperici Perforati, Herba Lamiophlomis, and Rhizoma Dioscoreae Nipponicae. The results show that Herba Hy- perici Perforati is Xing-han, Wei-xin ku （bitter）, attributing to Gan-jing, Shen-jing and Fei-jing, Herba Lamiophlomis attributing to Gan-jing and Pi-jing, and Rhizoma Dioscoreae Nipponicae attributing to Fei-jing, Dan-jing, and Gan-jing. Data mining technology is apparently a useful method for improving people＇s understanding of NCM.