The Construction Of Data Warehouse And Data Mining Based On HIS

With the continual development of the technology database , it makes it possible to store data information of magnanimity. But there are a phenomenon which the data are abundant ,but knowledge is poor When we face to the more and more super databases that expand rapidly . People don't set about to understand database , and to attain valuable information more difficultly At present, HIS has already been developed maturely in our country,and accumulated a large amount of data, but turned it into valuable data and guided hospital leaders and the doctors have few .Abased on this kind of current situation, this paper is an research about current HIS,and propose a scheme which sets up the data warehouse based on HIS and a kind of analyzing tool about data mining. Data warehouse is a subject-oriented,integrated,time-variant,and nonvolatile collection of data in support of management's decision making process.There are a bigger difference between it and traditional database This scheme mainly utilizes a characteristic of the data warehouse to sort out and gather by clearing up the business data produced in HIS,and store them to the data warehouse organically, so that the multidimension inquiry and query of the data.Meanwhile,data mining is one of the most forward position of research directions about database and information decision in the world at present, causes extensive concern in academia and industrial circle, develops various projects in the field, its main goal is to develop relevant methodology, theory and tool, in order to extract useful knowledge and mode from a large amount of data. This project abased on data warehouse is to utilize data mining to build Various difficult kinds of analyzing model, such as concept description and mining association rules,etc., thus offered the important decision for the doctor's guidance and certain grounds for the treatment of the disease and keep away.This paper is an extension of research on concept description and mining association rules, and make some improvements on existing Apriori algorithm,using QAR_SQL algorithm.concretely our main works are:Put forword a data warehouse model based on HIS build by of setting up.Have a research and discussion over various kinds of data mining algorithms, focusing on concept description and mining association rules, and make some improvements on existing Apriori algorithm, an extension of research on the existing mining association rules.Have a research and discussion over discretization methodology of continuous attributes using language field.Research to utilize a kind of standard SQL language to mine quantitative association rule --QAR _ SQL algorithm; Put forward to utilize abundant factor LS to appraise to the rule, LS can be used for producing the rule directly with threshold value of the credibility as threshold value too, and to LS <1. Length being 2 can produce association rule with negative word;...