A MODIFIED CLUSTERING ALGORITHM ITS APPLICATION IN DIGITIZATION OF TRADITIONAL CHINESE MEDICINE

摘要

The tongue intelligent diagnosis and inference system of Traditional Chinese Medicine (TCM) is a big, complex one, its data is of great amount and many types, also the data cluster has uncertainty. The clustering analysis can solve these difficult problems. In this article, the author introduced an advanced hybrid algorithm for artificial immune clustering and Radial-Basis Function (RBF) neural networks, by analyzing the weaknesses of the former artificial immune clustering algorithm. The algorithm based on artificial immune clustering was applied in Tongue Diagnosis (TD) of TCM (TCMTD) and the diagnosis model was constructed. It used the hepatic disease symptom as simulation. The experimental result demonstrated that the advanced artificial immune clustering algorithm could fleetly cluster on the large data and high dimension sample data and then determine optimal clustering centers to ensure the RBF neural network had good generalization ability, and the TCMTD model had good diagnostic ability, fast convergence rate and good generalization ability. So the modified clustering algorithm in digitization of TCMTD was feasible and valid.