Data-driven constructive induction in AQ17-PRE: A method and experiments

Author

Bloedorn, Eric ; Michalski, Ryszard S.

Author_Institution

Center for Artificial Intelligence, George Mason Univ., Fairfax, VA, USA

fYear

1991

fDate

10-13 Nov 1991

Firstpage

30

Lastpage

37

Abstract

A method is presented for constructive induction, in which new attributes are constructed as various functions of original attributes. Such a method is called data-driven constructive induction, because new attributes are derived from an analysis of the data (examples) rather than the generated rules. Attribute construction and rule generation are repeated until a termination condition, such as the satisfaction of a rule quality measure, is met. The first step of this method, the generation of new attributes, has been implemented in AQ17-PRE. Initial experiments with AQ17-PRE have shown that it leads to an improvement of the learned rules in terms of both their simplicity and their accuracy on testing examples

Keywords

inference mechanisms; knowledge based systems; learning systems; AQ17-PRE; attribute construction; data-driven constructive induction; rule generation; rule quality measure; Arithmetic; Artificial intelligence; Data analysis; Decision trees; Induction generators; Performance evaluation; Testing;

fLanguage

English

Publisher

ieee

Conference_Titel

Tools for Artificial Intelligence, 1991. TAI '91., Third International Conference on

Conference_Location

San Jose, CA

Print_ISBN

0-8186-2300-4

Type

conf

DOI

10.1109/TAI.1991.167073

Filename

167073