Font Size: a A A

Research And Application Of Data Mining Based On Rough Set

Posted on:2008-10-10Degree:MasterType:Thesis
Country:ChinaCandidate:N N LiFull Text:PDF
GTID:2178360278953538Subject:Computer software and theory
Abstract/Summary:PDF Full Text Request
Rough set(RS) is a new and important mathematical tool to deal with vagueness and imprecise problems, which was introduced by Pawlak Z in 1982. RS doesn't depend on additional information beyond the data set, so it's a potent tool for dealing with vague, imprecise, incomplete and uncertain data, and is a new technology in data mining.The process of data mining based on RS consists of data preparation, attribute reduction, the rules of formation and decision support. This paper is focused on attribute reduction and introduces the discretization of consistant attributes. Attribute reduction is one of the most important links during data mining based on RS. The correlative methods are divided into two kinds, selection before and deletion after. The former often starts with the computation of core, so this paper makes a study for the algorithms of core. When there exits lots of redundancy and conflicts, although those algorithms before can deal with both the consistent and inconsistent data and get the right results, they either delete the conflicts directly or reserve the conflicts completely instead of studying the information inside the data self. This paper proposes a weighted method to computer the core, which according to the degree of redundancy and conflict, expresses the references and reliability of the results. The experiment has shown the better results which are more near to the fact and improved the mode computing only from the algorithms. Discretization is the early work in data mining, the paper introduces the methods of discretization and discusses some related mothods such as the one based on the importance of each attribute and the one based on PSO, which give the theory guidance to the application.Finally, based on the experiment in the mushroom database, it proves the effectiveness of RS in applications. Because of the comparability between the mushroom data and those data in the national economy mobilization potential analyzation systems, the paper tries to apply the process to the latter and achieve the purpose of national economic mobilization.
Keywords/Search Tags:Rough Set, Data Mining, Attribute Reduction, Discretization
PDF Full Text Request
Related items