The Community for Technology Leaders
Green Image
ABSTRACT
Some machine learning algorithms enable the learner to extend its vocabulary with new terms if, for a given a set of training examples, the learner's vocabulary is too restricted for solving the learning task. In this article, the authors propose a filter that selects the potentially relevant terms from the set of constructed terms, and eliminates the terms which are irrelevant for the learning task. By biasing constructive induction (or predicate invention) to relevant terms only, the explored space of constructed terms can be much larger. The elimination of irrelevant terms is especially well suited for learners of large time or space complexity (such as genetic algorithms and neural nets). This article presents the Reduce algorithm for eliminating irrelevant terms and a case study in which the authors use Reduce to preprocess data for a hybrid genetic algorithm RL-ICET.
INDEX TERMS
CITATION
Nada Lavrac, Peter Turney, Dragan Gamberger, "A Relevancy Filter for Constructive Induction", IEEE Intelligent Systems, vol. 13, no. , pp. 50-56, March/April 1998, doi:10.1109/5254.671092
106 ms
(Ver 3.3 (11022016))