Concept Learning and Heuristic Classification in WeakTtheory Domains

Abstract This paper describes a successful approach to concept learning for heuristic classification. Almost all current programs for this task create or use explicit, abstract generalizations. These programs are largely ineffective for domains with weak or intractable theories. An exemplar-based approach is suitable for domains with inadequate theories but raises two additional problems: determining similarity and indexing exemplars. Our approach extends the exemplar-based approach with solutions to these problems. An implementation of our approach, called Protos, has been applied to the domain of clinical audiology. After reasonable training, Protos achieved a competence level equaling that of human experts and far surpassing that of other machine learning programs. Additionally, an “ablation study” has identified the aspects of Protos that are primarily responsible for its success.

[1]  K. Holyoak,et al.  The role of contradictions and counterexamples in the rejection of false sentences , 1975 .

[2]  Roger C. Schank,et al.  Dynamic memory - a theory of reminding and learning in computers and people , 1983 .

[3]  E. R. Bareiss,et al.  Gaining Autonomy During Knowledge Acquisition , 1989 .

[4]  G. Collins,et al.  Transcending inductive category formation in learning , 1986, Behavioral and Brain Sciences.

[5]  Michael R. Lowry,et al.  Learning Physical Descriptions From Functional Definitions, Examples, and Precedents , 1983, AAAI.

[6]  Thomas G. Dietterich,et al.  Selecting Appropriate Representations for Learning from Examples , 1986, AAAI.

[7]  D. Medin,et al.  The role of theories in conceptual coherence. , 1985, Psychological review.

[8]  Janet L. Kolodner,et al.  Extending Problem Solver Capabilities Through Case-Based Inference , 1987 .

[9]  Kalev Ruberg,et al.  House calls: building and maintaining a rule-base , 1989 .

[10]  Casimir A. Kulikowski,et al.  A Model-Based Method for Computer-Aided Medical Decision-Making , 1978, Artif. Intell..

[11]  L. Wittgenstein Philosophical investigations = Philosophische Untersuchungen , 1958 .

[12]  Michael Brady,et al.  Generating and Generalizing Models of Visual Objects , 1987, Artif. Intell..

[13]  E. Rosch,et al.  Family resemblances: Studies in the internal structure of categories , 1975, Cognitive Psychology.

[14]  Johanna D. Moore,et al.  A Reactive Approach to Explanation , 1989, IJCAI.

[15]  Patrick Henry Winston,et al.  Learning physical descriptions from functional definitions , 1983, AAAI 1983.

[16]  Butler W. Lampson,et al.  Annual Review of Computer Science , 1986 .

[17]  William J. Clancey,et al.  Heuristic Classification , 1986, Artif. Intell..

[18]  Douglas L. Medin,et al.  Context theory of classification learning. , 1978 .

[19]  Eleanor Rosch,et al.  Principles of Categorization , 1978 .

[20]  Edward E. Smith,et al.  Categories and concepts , 1984 .

[21]  Tom Michael Mitchell,et al.  Explanation-based generalization: A unifying view , 1986 .

[22]  S MalloryRichard Sources of Classification Accuracy in Protos , 1989 .

[23]  John H. Boose,et al.  Personal Construct Theory and the Transfer of Human Expertise , 1984, AAAI.

[24]  Bruce W. Porter,et al.  Developing a Tool for Knowledge Integration: Initial Results , 1990, Int. J. Man Mach. Stud..

[25]  Ryszard S. Michalski,et al.  A Theory and Methodology of Inductive Learning , 1983, Artificial Intelligence.

[26]  Adele E. Howe,et al.  How evaluation guides AI research , 1988 .

[27]  Jonathan Amsterdam,et al.  Some Philosophical Problems with Formal Learning Theory , 1988, AAAI.

[28]  Douglas L. Medin,et al.  Relationships between item and category learning: Evidence that abstraction is not automatic , 1983 .

[29]  A. Tversky,et al.  Judgment under Uncertainty: Heuristics and Biases , 1974, Science.

[30]  R. Bareiss,et al.  Supporting Start-to-Finish Development of Knowledge Bases , 2005, Machine Learning.

[31]  R. Bareiss Exemplar-Based Knowledge Acquisition , 1989 .

[32]  J. S. Aikins,et al.  A Representation Scheme Using Both Frames and Rules , 2022 .

[33]  Paul J. Feltovich,et al.  LCS: The role and development of medivcal knowledge in diagnostic expertise , 1984 .

[34]  Kenneth S Murray KI: An Experiment in Automating Knowledge Integr ation , 1988 .

[35]  Rita T Duran Concept Learning with Incomplete Data Sets (Master''s Thesis) , 1988 .

[36]  Patrick Henry Winston,et al.  Learning structural descriptions from examples , 1970 .

[37]  Bruce W. Porter,et al.  AI Research in the Context of a Multifunctional Knowledge Base: The BotanyKnowledge Base Project , 1988 .

[38]  Daniel L. Dvorak,et al.  Guide to CL-Protos: An Exemplar-based Learning Apprentice, , 1988 .

[39]  Edward A. Feigenbaum,et al.  Signal-to-Symbol Transformation: HASP/SIAP Case Study , 1982, AI Mag..

[40]  Jr. Robert Lee Simpson,et al.  A computer model of case-based reasoning in problem solving: an investigation in the domain of dispute mediation (analogy, machine learning, conceptual memory) , 1985 .

[41]  Edward E. Smith,et al.  Concepts and concept formation. , 1984, Annual review of psychology.

[42]  E. Shortliffe Mycin: computer-based medical consultations , 1976 .

[43]  Alen D. Shapiro,et al.  Structured induction in expert systems , 1987 .

[44]  Janet L. Kolodner,et al.  Maintaining Organization in a Dynamic Long-Term Memory , 1983, Cogn. Sci..

[45]  Peter Clark A Comparison of Rule and Exemplar-Based Learning Systems , 1990 .

[46]  David W. Aha,et al.  Learning Representative Exemplars of Concepts: An Initial Case Study , 1987 .

[47]  Janet L. Kolodner,et al.  Using Experience in Clinical Problem Solving: Introduction and Framework , 1987, IEEE Transactions on Systems, Man, and Cybernetics.