Based On Rough Set Data Mining Method

Rough Set theory is a mathematics tool for processing vague and imprecision knowledge, which is putted forward by 1982 from Poland scientist Z.Pawlak. The research object is represented in the form of information system by Rough set theory, the data table form commonly, each line of the table represents a research object, every column represents a kind of attribute. With 20 years development, the rough set theory not only in the theoretical study of continuously improve itself, but also have been successfully used in other fields, such as machine learning, image processing, expert systems, Pattern recognition, knowledge discovery in databases, medical diagnosis, financial data analysis etc.This article introduces data mining background and the current research to explore the concepts of data mining, working steps and key technologies. Also introduced the basic theory of rough set, an important content of rough sets as attribute reduction and rough set theory in data mining application-data attribute reduction method. Attribute reduction of rough set theory is an important part, most of algorithms currently available are based on complete information systems. Inconsistent information systems and data for the dynamic change are rarely considered. This paper proposed an improved algorithms in incomplete information system data in the context of dynamic changes of attribute reduction for the existing algorithms.Rough set applied to text classification, introduced the basic contents and methods in text classification, proposed methods based on tolerance rough set of text classification, based on the importance of attributes in attribute reduction algorithms are classification rules, which used in text classification, obtained the effectiveness and advantages of this method.