Research And Application Of KDD In Traditional Chinese Medical Diagnosis

Knowledge discovery in databases (KDD) is a rapidly emerging research field relevant to artificial intelligence and database system. KDD is the nontrivial process of identifying valid, novel, potentially useful, and ultimately understandable patterns in databases. In today's digital society, the explosive growth of many business, government, and scientific databases have far outpaced our ability to interpret and digest all the data, creating a need for a new generation of tools and techniques for automatic and intelligent database analysis, and that is the goal of KDD.Traditional Chinese Medicine (TCM) diagnosis has a very long history and is based on the study of a huge amount of valuable diagnosis information and documents, which are composed of ambiguous words with overloaded details. How to use them to serve for modern TCM diagnosis is a big challenge. The huge amount of information of TCM is just suited well for the application of KDD. This thesis is just aimed at discovering the hidden knowledge and extracting diagnostic rules in the TCM Diagnosis by applying KDD tools to medical diagnosis databases.In this thesis the characteristics of TCM diagnosis have been introduced and the basic concepts and some fundamental theories of KDD have been described thoroughly. According to the main task of our research, we have chosen some certain effective classification tools from various KDD techniques to satisfied the demands of TCM diagnosis. The following three tools, including Bayesian Classifier, Artificial Neural Networks (ANN) and Rough Sets (RS), have been studied especially and three classifier models have been given respectively. Several typical classification algorithms have been adopted and have been discussed in particular in the process of KDD. With further analysis of those classifiers and with the improvement of those algorithms, those tools, which have been proved useful, can be applied to the TCM databases more practically and be utilized more effectively.Also in this thesis an analysis and research work has been done to establish a whole TCM-KDD system, that is, the Chinese Medical Diagnosis Information System (CMIS). The structure and basic functions of this KDD system has been described and some key principles for constructing CMIS have been brought forward. The establishment and the improvement of this KDD system will benefit the development of modern TCM. Based on the results of our research, finally, some applications of KDD to the traditional medical diagnosis Expert System(ES) have been proposed.