Application of rough set to data mining: An Overview
ZHAO Wei dong 1, LI Qi hao 2 (1.School of Management, Fudan University, Shanghai 200433, China; 2.College of Mechanics and Automobile, Hefei University of Technology, Hefei 230009, China)
Data mining from large databases provides necessary support for decision making. As a new soft computing method, rough set theory is effective in dealing with indefinite, incomplete data and information. Because of its advantages over and integration with other soft computing methods, rough set has been widely used in many domains, especially in data mining (DM) , which is a hot topic . In this paper, some important issues including reduction of decision tables, incomplete information, discretization of continuously valued attributes, incremental data mining based on rough set and hybrid of rough set with other soft computing methods, such as neural networks, genetic algorithms, information theory, fuzzy set theory, are discussed in detail, which stand for some directions for further research. Also,the problems existing in them are pointed out for further research. A perspective is also proposed for development of rough set theory and its application.
【CateGory Index】： TP311.12