Based On Data Mining Of Hypertension Both Yin And Yang Deficiency Syndrome Evidence Law Research

Objective: Customs clearance to establish data mining based on decision tree and neural network, the method of reading ancient books basis and modern clinical typical cases of hypertension both Yin and Yang deficiency syndrome, syndrome elements such as information, establish discriminant model of hypertension both Yin and Yang deficiency syndrome; Through the inheritance of traditional Chinese medicine auxiliary system(V2.1) interpretation of hypertension both Yin and Yang deficiency syndrome characteristic information, the formulas for clinical medicine have guiding significance.Methords: 1. The base on hypertension of Yin and Yang deficiency syndrome, on the basis of ancient and modern basis, two syndrome elements database is set up respectively and their prescriptions database; The information will be collected by the four diagnosis and prescription information language standardization processing; To the standardization of information using binary variables 0, 1 assignment, the assignment 1 cases syndromes elements, does not appear and the lack of such as data assignment of 0; 2. Using SPSS 20.0 software of CHAID, CRT, QUEST and SPSS Modeler 14.1 C5.0 algorithm of syndrome elements analysis and decision tree model, for hypertension both Yin and Yang deficiency syndrome objective research provided the basis; Use SPSS 20.0 multilayer perceptron neural network and radial basis function(RBF) neural network to establish hypertension both Yin and Yang deficiency syndrome model; 3.Use of the inheritance of traditional Chinese medicine auxiliary system(V2.1) to analyze the data of the formulas. Statistics of hypertension both Yin and Yang deficiency syndrome drugs appear frequency, and analyze the hypertension of Yin and Yang deficiency syndrome two drugs of four gas, mixed and return. Using the principle of formula is selected five pairs, and based on the correlation analysis. Using the software of "improved mutual information method" to analyze correlation between traditional Chinese medicine, the complex system entropy clustering, sifting 3-4 flavour is the core of drug combination, and analyzes the possible new prescription, further guide clinical drug use.Results: 1. Establish hypertension both Yin and Yang deficiency syndrome database information and characterization of the formulas. 2. CHAID, CRT, QUEST and C5.0 algorithm is adopted to establish the hypertension both Yin and Yang deficiency syndrome diagnosis model and the accuracy were 93.1%、91.5%、91.5%、96.03%; Using multilayer perceptron neural network and radial basis function neural network to establish hypertension both Yin and Yang deficiency syndrome diagnosis model and the accuracy were 93.9%, 96.3%, the results are conform to the requirements of the statistical discrimination. 3. Use "of Chinese auxiliary system"(V2.1) analysis tools for hypertension in both Yin and Yang deficiency syndrome commonly used prescription and drugs were analyzed, and get five classes to tonify deficiency in medicine, medicine, liver extinguish wind medicine; More warm, grape, into the kidney, liver meridian; And through improved mutual information method and entropy clustering method based on complex system of statistical analysis of the core portfolio and New Zealand.Conclusions:This research through the establishment of hypertension both Yin and Yang deficiency syndrome element and attribute database, the formulas by using decision tree, neural network data mining methods to analyze the hypertension of both Yin and Yang deficiency syndrome distribution; Using "heritage of traditional Chinese medicine auxiliary system"(V2.1) analysis of hypertension both Yin and Yang deficiency syndrome using characteristic, the formulas to build specification syndromes, system, the formulas for the disease in clinical treatment has certain guiding significance.