Title :
Study on discretization based on rough set theory
Author :
Dai, Jian-Hua ; Li, Yuan-xiang
Author_Institution :
State Key Lab. of Software Eng., Wuhan Univ., China
Abstract :
Discretization of attributes with real values is an important problem in data mining based on rough set. And discretization based on rough set has some particular characteristics. The method of discretization based on rough set and Boolean reasoning is discussed. Determination of. candidate cuts is discussed in detail. A theorem is proposed. to show that all bound cuts can discern the same objects pairs as the whole initial cuts. A strategy to select candidate cuts is proposed based on the theorem. Under the strategy, the space complexity and time complexity of improved algorithm decline obviously. The experiments results also confirm that.
Keywords :
Boolean functions; computational complexity; data mining; rough set theory; Boolean reasoning; attribute discretization; bound cuts; candidate cut determination; data mining; object pairs; rough set theory; space complexity; time complexity; Computer science; Cybernetics; Data mining; Electronic mail; Laboratories; Machine learning; Mathematics; Set theory; Software engineering; Uncertainty;
Conference_Titel :
Machine Learning and Cybernetics, 2002. Proceedings. 2002 International Conference on
Print_ISBN :
0-7803-7508-4
DOI :
10.1109/ICMLC.2002.1167430