Data Classification Based On Rough Set And Its Application In Business Bank Supervision

Rough Set Theory, as a mathematic approach of analyzing data, was developed by Z.Pawlak in the 1980s.As a new data analysis approach, rough set theory can effectively deal with all kinds of inaccurate, inconsistent and incomplete data. It can find out useful information and rule in information system. Compared with other classic data classification approach, Rough Sets have many advantages. Since the 1980s, Rough Sets is flourishing in its mathematic theory, algorithm and application.This paper firstly systematically introduces Rough Set Theory's origin, advantages and its development. Mathematic modeling integrate flow using Rough Set is described in the paper.Discernibility matrix approach can induce all attribute reducts. But this approach will take much time in high dimension information system. This paper proposes the concept of indiscernibility matrix, then proves the relation between discernibility matrix and indiscernibility matrix and introduces an attribute reduction algorithm based on indiscernibility matrix. Compared with indicernibility matrix algorithm, this algorithm greatly reduces running time and memory space. There are many approach of decision rules generation, but many of them are faced to condition equivalence class. This paper proposes the concept of a-indiscernibility relation and produces a-decision class. On the basis, a-decision matrix is constructed, and decision rules generation algorithm and its incremental algorithm is proposed. Business bank supervision is composed of all kinds of relative qualitative and quantitative index. To analyze business banks' quality is equivalent to classify business bank . In the paper, Rough Set approach is used to provide a set of rules able to discriminate business bank rating.