论文部分内容阅读
目的:针对中药基本属性特征与中药功效之间呈现复杂的非线性不确定关系的问题,利用人工神经网络的模糊性及其良好的非线性拟合能力,挖掘中药复方药性特征与功效间的联系,实现中药复方功效的快速预测。方法:以126种补益类中药复方为样本分析对象,依据神经网络的方法及中药复方的组方原则,构建补益类中药复方功效与其性味归经之间关系的BP神经网络模型,对模型效果进行评价分析并运用MATLAB平台设计中药复方功效预测的GUI仿真。结果:本文所建立的模型在预测补益类中药复方的补气、补血、补阴、补阳功效准确率高达92.5%。结论:运用BP神经网络方法能够较好的模拟中药复方药性与功效之间的非线性映射,在预测补益类中药复方的功效上能达到很好的效果,为中药复方药性与功效研究提供新的技术支撑。
OBJECTIVE: To solve the problem of complex nonlinear uncertainties between the basic attributes of traditional Chinese medicine and the efficacy of traditional Chinese medicine (TCM), this paper uses the fuzziness of artificial neural network and its good non-linear fitting ability to explore the relationship between the medicinal properties and efficacy of traditional Chinese medicine , To achieve rapid prediction of the efficacy of traditional Chinese medicine compound. Methods: A total of 126 kinds of traditional Chinese medicine compound prescriptions were used as samples for analysis. According to the method of neural network and the principle of prescription of traditional Chinese medicine compound prescription, a BP neural network model was established to study the relationship between the compound prescription of Chinese herbs and its effect. Evaluate and analyze and use the MATLAB platform to design the GUI simulation of TCM compound efficacy prediction. Results: The model established in this paper can predict the effect of qi, blood, yin and yang of TCM compound prescriptions to 92.5%. Conclusion: The BP neural network method can better simulate the non-linear mapping between Chinese medicine compound prescription and efficacy, and can achieve good results in predicting the efficacy of traditional Chinese medicine compound prescription. It provides a new method for the study of herbal compound prescription and efficacy Technical Support.