A HYBRID RULE MINING APPROACH FOR A CLASSIFICATION PROBLEM-A CASE OF TWELVE-MERIDIAN DIAGONOSIS IN TRADITIAONL CHINESE MEDICINE

摘要

Data mining is the appliances for the valid new knowledge discovery from databases. A response model can be built as a decision model for the classification of a domain problem potential like expert systems. In this paper, a hybrid meta-evolutionary rule mining based approach is proposed to assess the total cholesterol data pattern detected from twelve meridians for extracting the decision rules including the predictors, the corresponding inequalities and parameters simultaneously so as to building a decision-making model with maximum classification accuracy. In real world, the diagnosis of traditional Chinese medicine through the twelve meridians is highly complicated in nature so that it’s hard to develop a comprehensive model. Recently, nonlinear and complex machine learning approaches such as neural networks and support vector machines have been demonstrated to be with more reliable than the conventional statistical approaches. The usefulness of using neural networks and support machines has been reported in literatures, but the most obstacles are in model building and use of model in which the classification rules are hard to be realized. Through the numerical experiment, we compared our results against the commercial data mining software, and then we show experimentally that the proposed approach is promising for improving prediction accuracy with fewer type II errors.