Study On Data Mining Model Based On Theory Of Granular Computing

Data mining is the process of discovering interesting knowledge from largeamounts of data stored either in databases, data warehouses, or other informationrepositories. In these many data mining tools, granular computing is an effectivemethod. Granular computing(GrC) is an umbrella term to cover any theories,methodologies, techniques and tools that make use of granules in problem solving.What is called a granule, is a clump of objects which are drawn together byindistingwishability, similarity or functionality. Granular computing may be studiedbased on two related issues, Granulation and Computation. The former deals with theconstruction, interpretation and representation of granules, and the latter deals withthe computing and reasoning with granules. Now, it has received many achievementsfor the research of granular computing, the research methods including rough set,fuzzy set and quotient space, etc. The foundation and development of rough setinfluenced and promoted largely for research and development of granular computing.After emergence of rough set, it was immediately found that it succeeded in dataclassification and knowledge reduct, and promptly reflected that it was more suitablefor researching these classification and reduct in Granular computing theory.Therefore, it soon become the hotspot in academic home and abroad.The author talked about granular computing theory from rough set perspective.First, the article reviewed classical rough set theory, and in the light of currentincomplete information systems, we presented a new rough set model for incompleteinformation systems. Next, the paper expatiated production of granular computingtheory. Finally, we put forward a data mining model based on granular computing,analyzed centrolly data pretreatment module, attributes reduct module and rulemining module, proposed an incompletion method for incomplete data, attributesreduct arithmetic based on granularity entropy, notion of user identified data miningand its implement arithmetic, an instance was appended to illuminate the whole datamining steps.