Learning Domain Theories using Abstract Beckground Knowledge

Substantial machine learning research has addressed the task of learning new knowledge given a (possibly incomplete or incorrect) domain theory, but leaves open the question of where such domain theories originate. In this paper we address the problem of constructing a domain theory from more general, abstract knowledge which may be available. The basis of our method is to first assume a structure for the target domain theory, and second to view background knowledge as constraints on components of that structure. This enables a focusing of search during learning, and also produces a domain theory which is explainable with respect to the background knowledge. We evaluate an instance of this methodology applied to the domain of economics, where background knowledge is represented as a qualitative model. This research was primarily conducted while at University of Ottawa, supported by NSERC strategic grants.

[1]  Larry A. Rendell,et al.  Feature construction: an analytic framework and an application to decision trees , 1990 .

[2]  Yves Kodratoff,et al.  Machine Learning — EWSL-91 , 1991, Lecture Notes in Computer Science.

[3]  Ranan B. Banerji,et al.  Learning theoretical terms , 1994, COLT 1994.

[4]  James C. Spohrer,et al.  Reasoning-Driven Memory Modification in the Economics Domain , 1984 .

[5]  Raymond J. Mooney,et al.  Constructive Induction in Theory Refinement , 1991, ML.

[6]  Raymond J. Mooney,et al.  First-Order Theory Revision , 1991, ML.

[7]  Paul Compton,et al.  Inductive knowledge acquisition: a case study , 1987 .

[8]  Anne v. d. L. Gardner,et al.  The Design of a Legal Analysis Program , 1983, AAAI.

[9]  Jerzy W. Bala,et al.  The Principal Axes Method for Constructive Induction , 1992, ML.

[10]  Igor Mozetič.  THE ROLE OF ABSTRACTIONS IN LEARNING QUALITATIVE MODELS , 1987 .

[11]  Francesco Bergadano,et al.  A Knowledge Intensive Approach to Concept Induction , 1988, ML Workshop.

[12]  Gerald DeJong.  Explanation-Based Learning with Plausible Inferencing , 1990 .

[13]  Peter Clark,et al.  Rule Induction with CN2: Some Recent Improvements , 1991, EWSL.

[14]  Kenneth D. Forbus.  Qualitative Process Theory , 1984, Artif. Intell..

[15]  James C. Lester,et al.  Generating Context-Sensitive Explanations in Interactive Knowledge-BasedSystems , 1991 .

[16]  North Hanover StreetGlasgow.  Relating Relational Learning Algorithms , 1992 .

[17]  Luc De Raedt,et al.  Constructive Induction by Analogy : a method to learn how to learn , 1989 .

[18]  Stephen Muggleton,et al.  Efficient Induction of Logic Programs , 1990, ALT.

[19]  Ray Bareiss,et al.  Concept Learning and Heuristic Classification in WeakTtheory Domains , 1990, Artif. Intell..