Knowledge Discovery And Reasoning Algorithm Study In Medical Diagnose Expert System

TCM emphasis dialectical treatment, TCM doctors perform treatment to patients reasonably based on the clinical symptoms, whose diagnostic methods mainly include the look,listen,question and feel the pulse four ways of diagnosis which is called “Four Diagnostic Methods”. TCM diagnosis system refers to the computer program system which can simulate the activity in the brains of the experts of TCM to analysis and process the symptoms, signs of the patients intelligently. Though the study has been conducted for decades, but due to the typical indeterminacy and fuzziness of clinical information in TCM diagnosis theory and its great amount of data etc., its progress is slow and restricted.In this paper, the practical clinical diagnosis and treatment of famous TCM medical record is the research object, by using soft sets association rules algorithm for knowledge discovery research, whose results are applied to constructing knowledge base of the system.This paper attempts to use the inclusion degree of soft sets algorithm and the theory in the formal concept applying to the knowledge discovery in TCM, and on this basis designing the TCM diagnostic system based on Internet technology and belief rule base.In this paper, the design and development of the TCM diagnosis system is finished in the environment of Visual Studio 2010 and adopts ASP.net as re-exploitation tool, using the JQuery?Css+Html?Ajax and other techniques combined with the simulation results of Matlab program and the algorithm of RIMER to achieve the full system. This system is consisted of knowledge discovery and diagnostic reasoning. By using the knowledge discovery algorithm about fuzzy big data to extract the association rules, which contains the inclusion degree and association rules between the attributes sets in soft sets concept.In order to ensure the correctness of the result of knowledge discovery, this paper takes advantage of the attribute partial order structure diagrams method to do the visual KDD on the same data, and compared the result with the knowledge discovery result of the approach to association rules mining using inclusion degree of soft sets. In the diagnostic reasoning part of the system, it uses the knowledge discovery results as previouslymentioned to construct the Belief-rule base in the RIMER reasoning algorithm, which is finally used for reasoning diagnosis and giving recommended prescription according to the patients' information.Although this system is intended for applying in scientific study and clinical auxiliary diagnosis, after further development, it can also be used in original data collection and analysis in the background of Internet and big data.