A Tentative Approach to Minimal Reducts by Combining Several Algorithms
Finding minimal reducts is a NP-hard problem. For obtain a feasible solution, depth-first-searching is mainly used and a feasible reduct always can be gotten. Whether the feasible reduct is a minimal reduct or not and how far it is to minimal reduct, both are not known. It only gives the information that how many attributes it has and it is a reduct. Based on rough sets reduction theory and the data structure of information system, the least condition attributes to describe the system’s classified characteristics can be known. So an area of searching minimal reducts is decided. By binary search in the area, the minimal reducts can be gotten quickly and doubtlessly.
Keywordsrough sets algorithm attribute reduction minimal reduct
Unable to display preview. Download preview PDF.
- 2.Pawlak, Z.: Rough Sets and Their Applications. Microcomputer Applications 13(2), 71–75 (1994)Google Scholar
- 4.Xu, N.: The Theory and Technique Research of Attribute Reduction in Data Mining Based on Rough Sets, PhD dissertation, Guangdong University of Technology (2005)Google Scholar
- 5.Ni, Z., Cai, J.: Discrete Mathematics. Science Publishes (2002)Google Scholar
- 6.Zhang, W., Wu, W., Liang, J., Li, D.: Theory and Method of Rough Sets. Science Publishes (2001)Google Scholar
- 7.Guo, J.: Rough set-based approach to data mining, PhD dissertation, Department of Electrical Engineering and Computer Science, Case Wester University, USA (2003)Google Scholar
- 8.Hu, X.: Knowledge Discovery in Database: An Attribute-oriented Rough Set Approach (Rules, Decision Matrices), PhD dissertation, The University of Regina, Canada (995)Google Scholar
- 10.Shi, Z.: Knowledge Discovery. Tsinghua University Press, Beijing (2002)Google Scholar
- 11.Duntsch, I., Gediga, G., Orlowska, E.: Relation Attribute Systems II: Reasoning with Relations in Information Structures. In: Peters, J.F., Skowron, A., Marek, V.W., Orłowska, E., Słowiński, R., Ziarko, W. (eds.) Transactions on Rough Sets VII. LNCS, vol. 4400, pp. 16–35. Springer, Heidelberg (2007)CrossRefGoogle Scholar