A Computerized Diagnostic Model Based on Naive Bayesian Classifier in Traditional Chinese Medicine

Traditional Chinese Medicine (TCM) is one of the most important complementary and alternative medicines. In this paper, a novel computerized diagnostic model is proposed for promoting standardization and popularization of TCM diagnosis. Firstly, the symptoms are selected by learning Bayesian network structure from a database of cases incorporating with mutual information theory. The Markov blanket of the target variable in the structure is selected as symptom set. Secondly, the mapping relationships between the symptom set and diagnostic results are constructed based on naive Bayesian classifier. The model is used to make the quantitative diagnosis of apoplexy and obtains relative reliable predictions. The rate of predictive accuracy in diagnosing apoplexy reaches 96.7%. The results suggest that the model constructed is feasible and effective and can be expected to be useful in the modernization of TCM.