Learning Representation by Integrating Case-Based and Inductive Learning

Dennis Connolly, Steve Christey, Phyllis Koton, Stuart McAlpin, and Alice Mulvehill

This paper describes a machine learning approach combining case-based reasoning with inductive learning in order to learn representation of problem solving context. The essence of this approach is the use of conceptual clustering to both facilitate efficient retrieval of cases and to induce concepts representing generalizations of context. In addition, learning by examples is used to acquire rules to guide adaptation of retrieved cases. The concepts induced via clustering are used to represent generalized context in the learning and application of acquired adaptation rules.


This page is copyrighted by AAAI. All rights reserved. Your use of this site constitutes acceptance of all of AAAI's terms and conditions and privacy policy.