Comparison of ID3 and its generalized version

Author

Hong, Tzung-Pei ; Tseng, Shian-Shyong

Author_Institution

Dept. of Comput. Sci., Chung-Hua Polytech. Inst., Hsin-Chu, Taiwan

fYear

1992

fDate

28-30 May 1992

Firstpage

241

Lastpage

244

Abstract

A generalized ID3 learning algorithm, which possesses more processing capabilities than the original ID3, is proposed. This generalized ID3 learning algorithm can manage uncertain training instances and consider the different importance of different training instances in finding an appropriate decision tree in a noisy environment. If appropriate priori domain knowledge is available, this algorithm can further utilize it in reducing the effect of noise. Relations on generality, accuracy, tree complexity and time complexity of ID3 and generalized ID3 are also discussed, with experiments verifying their correctness

Keywords

computational complexity; learning (artificial intelligence); learning systems; program verification; trees (mathematics); ID3 learning algorithm; correctness verification; decision tree; noisy environment; priori domain knowledge; time complexity; tree complexity; uncertain training; Computer science; Decision trees; Entropy; Environmental management; Information science; Learning systems; Management training; Noise reduction; Signal to noise ratio; Working environment noise;

fLanguage

English

Publisher

ieee

Conference_Titel

Computing and Information, 1992. Proceedings. ICCI '92., Fourth International Conference on

Conference_Location

Toronto, Ont.

Print_ISBN

0-8186-2812-X

Type

conf

DOI

10.1109/ICCI.1992.227664

Filename

227664