Research On Data Mining Technique Based On Data Element Standard And Rough Sets Theory

As the inevitable trend of the information-based development at present, the share of the information and work cooperation in different field have already become the hot spot of the current information-based foundation. Because they are on the basis of the coincide and accurate understanding of the meaning and expression of data share by the information users and owners, so it is not necessary to emphasize the importance of Data Standard. Data Standard aims at offering the solutions for the popular, easy, and practical information exchanging terrace, which also lays the foundation of mathematics for sharing, gathering information between the information system, and it draws more and more attentions in the international information field.Due to the wide application of modern computer-realized data collection and Database technology over past years, many different industries have already collected huge amount of data. As traditional data analyze methods are hardly to discover the inherit or implicit information. Data Mining whicjh widely applies Statistics theory .Rough Sets, Fuzzy theory .Machine Learning as well as Database and other scientific ways is essentially the nontrivial process of identifying valid, novel, potentially useful and ultimately understandable patterns in data°As one of technology theories applied by Data Mining, Rough Sets which mainly targets to identify incomplete and uncertain data, is first forwarded by Z Pawlak in 1982. Since it's having very strong qualitative analysis ability and deployments do not need much preexperienced knowledge, Compared with other Data Mining methods, Rough Sets can bring more convince to application fields.This thesis focuses on the research of Data Mining technique based on Data Element Standard and Rough Sets. First, on the basis of the theory research on Data Element Standard and XML, the thesis provides the design of the Mapping model between Data Element and XML Schema, which emphasizes the Mapping rule and the saving exchanging format of Data Element based on XML Schema, based on which founds the communication between Data Standard and Data Mining. Second, the thesis sheds light on the fact that Rough Sets is an efficient scientific way to identify incomplete and uncertain data and thus fits for Data Mining in database. This part focuses on the research on the base quality of Rough Sets and Attributes Reduction. Third, the thesis stresses to research Data Mining technique based on Data Element Standard and Rough Sets, then puts forward the Data Mining frame, which includes two parts, data preprocessing based on Data Element Standard and rule mining based on Rough Sets, and further puts forward the design of MetaData and the new attributes reduction method. In the end, the thesis proposes an Agriculture Data Mining instance based on the research before, and further discusses the application of Data Element Standard in Agriculture information-based development. What this thesis aims at is to found the communication between Data Element Standard and Rough Sets and Data Mining through analyzing the essence and process of them. And it will provide the new method for the richment and perfection of Data Mining system.