A comparative study on heuristic algorithms for generating fuzzy decision trees

Author

Wang, X.Z. ; Tsang, E.C.C. ; Yeung, D.S.

Author_Institution

Dept. of Comput., Hong Kong Polytech. Univ., Kowloon, Hong Kong

Volume

3

fYear

1999

fDate

6/21/1905 12:00:00 AM

Firstpage

343

Abstract

Fuzzy decision tree induction is an important way of learning from fuzzy examples. Since the construction of an optimal decision tree is NP-hard, heuristic algorithms are necessary. In the paper, three heuristic algorithms for generating fuzzy decision trees have been compared and analyzed. One of them was proposed by the authors. The analytic comparison is based on four issues: expanded attribute selection, leaf-node standard, matching approach and computation complexity. The comparative results have given some important information about heuristics and have shown useful guidelines to choose an appropriate heuristic for a particular problem

Keywords

computational complexity; decision trees; fuzzy logic; learning by example; NP-hard problem; computation complexity; expanded attribute selection; fuzzy decision tree induction; heuristic algorithms; leaf-node standard; matching approach; optimal decision tree; Algorithm design and analysis; Buildings; Classification tree analysis; Decision trees; Fuzzy sets; Guidelines; Heuristic algorithms; Induction generators; Information analysis; Tree graphs;

fLanguage

English

Publisher

ieee

Conference_Titel

Systems, Man, and Cybernetics, 1999. IEEE SMC '99 Conference Proceedings. 1999 IEEE International Conference on

Conference_Location

Tokyo

ISSN

1062-922X

Print_ISBN

0-7803-5731-0

Type

conf

DOI

10.1109/ICSMC.1999.823227

Filename

823227