Research On Attribute Reduction Algorithms Based On Rough Sets Theory

The rough set theory is a mathematics tool in processing inaccurate, inconsistent and incomplete problems, which can find the implicit knowledge and potential regulations by directly analyzing and deducing the data without any prior information except the data set. The attribute reduction is a important problem of the rough set theory, which delete the redundant attributes on the condition of keeping on the invariable classifying ability and the fast algorithm of reduction is one of the main research contents of the theory of rough set, which is a key step of knowledge acquisition. Therefore, attribute is focused on in this dissertation. The main content is divided as follows.(1) Relevant knowledge of the rough set theory is introduced and some classic attribute reduction algorithms are analyzed systematically.(2) On the existing attribute reduction algorithm, a modified attribute reduction algorithm is proposed , which need scan every attribute once in the decision table according to relative positive region concept in the rough set theory. It enable the attribute value simplify at the same time.(3) On the basis of elicitation information which is based on attribute significance, putting forward seeking nuclear and reduction method based on degree importance of attribute and degree of dependence. The method is different from the previous method. the third algorithm defined the attribute importance using the information entropy . The effectiveness and the feasibility of the algorithms in this paper are clearly demonstrated by the example analysis.The information system isn't a simplest one after attribute reduction .It includes some redundancy information and needs value reduction. The general algorithm and its improved algorithms are proposed in this paper.Using these algorithms we can reduce the complexity of the original attribute reduction algorithm and value reduction algorithm greatly and finally acquire the optimal decision-making rules of the information system.At last, the reduction algorithms are applied(used)to the method of data mining. The experiment results indicate this method is very efficient.