The Research On Data Reduction Based On The Theory Of Information Granules And Granular Computing

In the end of 1980s of the 20th century, people have realized that more and more accumulated information of information systems has come to the risk margin of information exploding. Reduction information systems mainly cut down unnecessary or redundancy information, but not affect their original functions. Granular computing(GrC) theory,introduced by American scientist L.A.Zadeh------the father of fuzzy set theory------in 1970s ofthe 20 century, is used to solve the problem of information discontinuity in real world. Nowadays, granular computing theory has already become a research focus in rough set theory. It will act an important role in the application of rough set theory to data reduction.This paper mainly discusses rough set theory and granular computing theory. We also do some meaningful research on the application of rough set theory and granular computing theory to data reduction. The creativity of this paper lies in two points. First, we propose data reduction algorithms based on the binary granular computing theory. Second, we obtained decision rules throw data reduction based on granular computing theory in a foreign classic car database.This paper can be divided into five parts. The first part discusses the conception of data mining, the process of data mining, skills and tools usually used in data mining, the research focuses of data mining nowadays and developing direction in future . The second part discusses the theory of rough set, the generalized model of rough set and the developing status of rough set. The third part discusses the basic conception of granular computing, computing with words , granular computing of rough set theory, granular reasoning, the expression and computation of binary granules and the current status and prospect of granular computing. The fourth part discusses the data reduction based on granular computing, we propose algorithms about data reduction based on binary granular computing, and briefly depict nodus in binary granular computing. The fifth part discusses the application of data reduction based on granular computing.