Research On Knowledge Representation And Reasoning Based On Deductive Lattice In Clinical Decision Support System

The research of clinical decision support system began in the 1960 s.With the gradual maturity of CDSS technology,it has become more and more widely used in the field of medical auxiliary diagnosis.As the scale of the system becomes more and more complex,the current technology of building CDSS knowledge base and inference engine presents certain problems including not being able to deal with large-scale knowledge,low efficiency of reasoning and difficulty in acquiring and updating knowledge.The main reason is that the design of knowledge representation and reasoning algorithms has always been a difficult point of CDSS.Therefore,the research on knowledge representation and reasoning algorithms in CDSS can help to further improve and develop the technology in this field.This thesis studies the main techniques applied in CDSS based on the research of CDSS.Then in order to solve the problem put forward above,in this thesis,a method of constructing the deductive lattice based inference is presented.A method modular and object-oriented medical knowledge base is presented to reduce the coupling degree of knowledge and the difficulty of acquiring knowledge.A method based on deductive lattice is proposed to improve efficiency of reasoning.We provide a new idea to build an inference engine for CDSS.The thesis first introduces the significance of current CDSS research,briefly analyzes the CDSS and the current research status of CDSS,and analyzes knowledgebased and non-knowledge-based CDSS.Then conduct in-depth research on the main methods used by knowledge-based CDSS.Then analyze the ideas and characteristics of these methods,point out its structure and the advantages and disadvantages of these methods.And explains why there are such problems.The thesis introduces the knowledge representation and reasoning algorithms based on the deductive lattice.Then introduce the inference structure of the deductive lattice and its inference algorithm,and proposes a reasoning process based on humancomputer interaction.The improvement of the method for the production rule is summarized and analyzed.