Research On Incremental Attribute Reduction Based On Rough Set

Rough set theory is an effective method of data mining.It is being recognized gradually. Its basic theory is utilizing equivalence relation class,through reduction,obtaining knowledge and reduction of knowledge with the same ability of classification.At first,this paper describes the basic principles and the main methods of data mining and rough set theory.Then it studies essential methods and algorithms of attribute reduction. And a new attributes reduction algorithm based on information entropy is proposed to tackle the problems involved in rough set. The new algorithm adopts information entropy as attribute selecting criterion. In order to meet dynamic database, this paper proposes a new incremental attributes reduction algorithm based on information entropy by modifying this method. At last the definition of feature matrix is extended,based on which an incremental rule extraction algorithm is proposed.