An approach to attribute generalization in incomplete information systems

Publication Type:
Conference Proceeding
Proceedings of International Conference on Machine Learning and Cybernetics, 2003, 2003, 3 pp. 1698 - 1703
Issue Date:
Full metadata record
Files in This Item:
Filename Description SizeFormat
2009005269OK.pdf291.74 kBAdobe PDF
Data mining is the efficient discovery of previously unknown patterns in large databases. How to use the existing knowledge to update knowledge is one of the important research areas in data mining. Since knowledge found by rough set is an apparent quantitative description and can be understood, of which data mining is in pursuit. Presently, several approaches based on classical rough set which aims at complete information system have been proposed for the mining task of updating knowledge. However, many information systems are incomplete in practical application. So it is important to develop approaches for updating knowledge in incomplete information systems in order to support more effective data mining. In this paper, based on an extension of the classical rough set theory for dealing with incomplete information systems, we have proposed a method for incremental updating approximations of a concept in incomplete information systems which may realize adding and deleting some attributes simultaneously at a time, that is very important to effectively handle dynamic attribute generalization and enhance the efficiency of data mining.
Please use this identifier to cite or link to this item: