Learning the Classic Description Logic: Theoretical and Experimental Results

Abstract We present a series of theoretical and experimental results on the learnability of description logics. We first extend previous formal learnability results on simple description logics to C-CLASSIC, a description logic expressive enough to be practically useful. We then experimentally evaluate two extensions of a learning algorithm suggested by the formal analysis. The first extension learns C-CLASSIC descriptions from individuals. (The formal results assume that examples are themselves descriptions.) The second extension learns disjunctions of C-CLASSIC descriptions from individuals. The experiments, which were conducted using several hundred target concepts from a number of domains, indicate that both extensions reliably learn complex natural concepts.

[1]  Michael J. Pazzani,et al.  Detecting and correcting errors in rule-based expert systems: an integration of empirical and explanation-based learning , 1991 .

[2]  Alexander Borgida,et al.  Computing Least Common Subsumers in Description Logics , 1992, AAAI.

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

[4]  Sunit K. Gala,et al.  Classification as a query processing technique in the CANDIDE semantic data model , 1989, [1989] Proceedings. Fifth International Conference on Data Engineering.

[5]  N. Littlestone.  Learning Quickly When Irrelevant Attributes Abound: A New Linear-Threshold Algorithm , 1987, 28th Annual Symposium on Foundations of Computer Science (sfcs 1987).

[6]  Jon R. Wright,et al.  A Knowledge-Based Configurator that Supports Sales, Engineering, and Manufacturing at AT&T Network Systems , 1993, AI Mag..

[7]  William W. Cohen,et al.  Learnability of description logics , 1992, COLT '92.

[8]  Deborah L. McGuinness,et al.  CLASSIC: a structural data model for objects , 1989, SIGMOD '89.

[9]  Premkumar T. Devanbu,et al.  LaSSIE—a knowledge-based software information system , 1991, ICSE '90.

[10]  William W. Cohen.  Compiling prior knowledge into an explicit basis , 1992, ICML 1992.

[11]  J. R. Quinlan.  Learning Logical Definitions from Relations , 1990 .

[12]  Leslie G. Valiant,et al.  A theory of the learnable , 1984, CACM.

[13]  Peter F. Patel-Schneider,et al.  Living wiht Classic: When and How to Use a KL-ONE-Like Language , 1991, Principles of Semantic Networks.

[14]  James G. Schmolze,et al.  The KL-ONE family , 1992 .

[15]  Peter F. Patel-Schneider,et al.  A Semantics and Complete Algorithm for Subsumption in the CLASSIC Description Logic , 1993, J. Artif. Intell. Res..

[16]  Robert M. MacGregor,et al.  The Evolving Technology of Classification-Based Knowledge Representation Systems , 1991, Principles of Semantic Networks.

[17]  David Haussler,et al.  Learnability and the Vapnik-Chervonenkis dimension , 1989, JACM.

[18]  Chidanand Apté,et al.  Organizing Knowledge in a Complex Financial Domain , 1987, IEEE Expert.

[19]  Thomas G. Dietterich,et al.  Learning and Inductive Inference , 1982 .