Learning Strategies and Automated Knowledge Acquisition

An Overview
  • Ryszard S. Michalski
Part of the Symbolic Computation book series (SYMBOLIC)

Abstract

Fundamental learning strategies are discussed in the context of knowledge acquisition for expert systems. These strategies reflect the type of inference performed by the learner on the input information in order to derive the desired knowledge. They include learning from instruction, learning by deduction, learning by analogy and learning by induction. Special attention is given to two basic types of learning by induction: learning from examples (concept acquisition) and learning from observation (concept formation without teacher). A specific form of learning from observation, namely, conceptual clustering, is discussed in detail, and illustrated by an example. Conceptual clustering is a process of structuring given observations into a hierarchy of conceptual categories.

An inductive learning system generates knowledge by drawing inductive inferences from the given facts under the guidance of background knowledge. The background knowledge contains previously learned concepts, goals of learning, the criteria for evaluating hypotheses from the viewpoint of these goals, the properties of attributes and relations used to chracterize observed events, and various inference rules for transforming concepts or expressing them at different levels of abstraction.

Keywords

Explosive Dition Univer Rote 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Boulanger, A: The Expert System PLANT/cd: A Case Study in Applying the General Purpose Inference System ADVISE to Predicting Black Cutworm Damage in Corn, M. S. Thesis, UIUCDS-R-83–1134, Department of Computer Science, University of Illinois, Urbana, IL, July 1983.Google Scholar
  2. Buchanan BG, Mitchell TM, Smith RG, Johnson CR Jr., Models of Learning Systems, Tech Rept STAN-CS-79–692, Computer Science Department, Stanford University, 1979.Google Scholar
  3. Burstein MH: Concept Formation by Incremental Analogical Reasoning and Debugging, Proceedings of the 2nd International Machine Learning Workshop, Allerton House, University of Illinois at Urbana-Champaign, June 22–24, 1983.Google Scholar
  4. Carbonell JG, Learning by Analogy: Formulating and Generalizing Plans From Past Experience, chapter in Machine Learning: An Artificial Intelligence Approach, Michalski RS, Carbonell J, Mitchell T (eds.), Tioga Publishing Company, 1983.Google Scholar
  5. Carbonell JG, Derivational Analogy: A Theory of Reconstructive Problem Solving and Expertise Acquisition, chapter in Machine Learning, Volume II, Morgan Kaufman Publishers, Inc., 1986.Google Scholar
  6. Chilausky R, Jacobsen B, Michalski RS: An Application of Variable-Valued Logic to Inductive Learning of Plant Disease Diagnostic Rules, Proceedings of the 1976 International Symposium on Multiple-Valued Logic, Utah State University, Logan, Utah, May 25–28, 1976.Google Scholar
  7. Davis R, Knowledge Acquisition in Rule Based Systems: Knowledge About Representation as a Basis For System Construction and Maintenance. Chapter in Pattern-Directed Inference Systems, Waterman DA, Hayes-Roth F (eds.), Academic Press, 1978.Google Scholar
  8. DeJong G, Mooney R, Explantation-Based Learning: An Alternative Review, Machine Learning Journal, Vol I, No 2, pp 145–176, Kluwer Academic Publishers, 1986.Google Scholar
  9. Dietterich TG, Michalski RS, A Comparative Review of Selected Methods for Learning From Examples, chapter in Machine Learning: An Artificial Intelligence Approach, Michalski RS, Carbonell J, Mitchell T, (eds.), Tioga Publishing Company, 1983.Google Scholar
  10. Doyle RJ, Constructing and Refining Causal Explanations from an Inconsistent Domain Theory, Proceedings of AAAI-86, Fifth National Conference on Artificial Intelligence, Vol I, pp 538–544, Philadelphia, PA, August 1986.Google Scholar
  11. Hass N, Hendrix GG, Learning by Being Told: Acquiring Knowledge for Information Management, chapter in Machine Learning: An Artificial Intelligence Approach, Michalski RS, Carbonell J, Mitchell T, (eds.), Tioga Publishing Company, 1983.Google Scholar
  12. Hoff WA, Michalski RS, Stepp RE, Induce Z: A Program For Learning Structural Descriptions From Examples, Technical Report, Department of Computer Science, University of Illinois, Urbana, Illinois, September, 1983.Google Scholar
  13. Hu Chi-Heng, Pao Yoh-Han, Processing of Pattern Based Information, Part II: Description of Inductive Inference in Terms of Transition Networks, Technical Report 102–82 Part II, Case Institute of Technology, Case Western Reserve University, Cleveland, Ohio 44106, 1982.Google Scholar
  14. Langley P, Bradshaw GL, Simon HA, Rediscovering Chemistry With the BACON System, chapter in Machine Learning: An Artificial Intelligence Approach, Michalski RS, Carbonell J, Mitchell T, (eds.), Tioga Publishing Company, 1983.Google Scholar
  15. Lebowitz M, Not the Path to Perdition: The Utility of Similarity-Based Learning, Proceedings of AAAI-86, Fifth National Conference on Artificial Intelligence, Vol I, pp 533–537, Philadelphia, PA, August, 1986.Google Scholar
  16. Lenat D, AM An Artificial Intelligence Approach to Discovery in Mathematics as Heuristic Search, Computer Science Department, Rept STAN-CS-76–750, Stanford University, Stanford, CA, 1976.Google Scholar
  17. Lenat D, Hayes-Roth F, Klahr PH, Cognitive Economy in a Fluid Task Environment, Proceedings of the International Machine Learning Workshop, Michalski RS (ed.), Allerton House, June 22–24, 1983.Google Scholar
  18. Medin DL, Structural Principles in Categorization, Developments of Perception and Cognition, Shepp B, Tighe T (eds.), Erlbaum, Hillsdale, NJ, pp 203–230, 1983.Google Scholar
  19. Medin DL, Wattenmaker WS, Michalski RS, Constraints and Preferences in Inductive Learning: An Experimental Study Comparing Human and Machine Performance, ISG Report 86–1, Computer Science Department, University of Illinois, Urbana, January, 1986.Google Scholar
  20. Michalski RS, A Variable-Valued Logic System as Applied to Picture Description and Recognition, Chapter in the book Graphic Languages, Nake F, Rosenfeld A (eds.), North-Holland Publishing Co., 1972.Google Scholar
  21. Michalski RS, Knowledge Aquisition Through Conceptual Clustering: A Theoretical Framework and an Algorithm for Partitioning Data Into Conjunctive Concepts, Policy Analysis and Information Systems, Vol 4, No 3, pp 219–244, 1980b.MathSciNetGoogle Scholar
  22. Michalski RS, Chilausky RL, Knowledge Acquisition by Encoding Expert Rules Versus Computer Induction From Examples: A Case Study Involving Soybean Pathology, International Journal for Man-Machine Studies, No 12, pp 63–87, 1980.CrossRefGoogle Scholar
  23. Michalski RS, A Theory and Methodology of Inductive Learning, chapter in Machine Learning: An Artificial Intelligence Approach, Michalski RS, Carbonell JG, Mitchell TM (eds.), Tioga Publishing Company, 1983.Google Scholar
  24. Michalski RS, Ko H, Chen K, SPARC/E - An Eleusis Rule Generator and Game Player, UIUCDCS-F-85–941, ISG Report 85–11, Department of Computer Science, University of Illinois, Urbana, February 1985.Google Scholar
  25. Michalski RS, Stepp RE, Automated Construction of Classifications: Conceptual Clustering Versus Numerical Taxonomy, IEEE Trans. on Pattern Analysis and Machine Intelligence, Vol PAMI-5, No 4, pp 396–410, July, 1983 a.Google Scholar
  26. Michalski RS, Stepp RE, Learning From Observation: Conceptual Clustering, chapter in Machine Learning, Michalski RS, Carbonell J, Mitchell T (eds.), Tioga Publishing Company, 1983b.Google Scholar
  27. Michalski RS, Mozetic I, Hong J, Lavrac N, The Multi-Purpose Incremental Learning System AQ15 and Its Testing Application to Thtee Medical Domains, Proceedings of AAAI-86, Fifth National Conference on Artificial Intellingence, Vol 2, pp 1041–1045, Philadelphia, PA, August 1986.Google Scholar
  28. Michalski RS, Ko H, Chen K, Qualitative Prediction: A Method and Program SPARC/G, Chapter in Expert Systems, Guetler C (ed.), Academic Press, 1987.Google Scholar
  29. Mitchell TM, Version Spaces: An Approach to Concept Learning, Thesis PhD, Stanford University, Stanford, CA 1978.Google Scholar
  30. Mitchell TM, Utgoff PE, Banerji R, Learning by Experimentation: Acquiring and Refining Problem-Solving Heuristics, chapter in Machine Learning: An Artificial Intelligence Approach, Michalski RS, Carbonell JG, Mitchell TM (eds.), Tioga Publishing Company, 1983.Google Scholar
  31. Mitchell TM, Keller RM, Kedar-Cabelli ST, Explanation-based Generalization: A Unifying View, Machine Learning, pp 47–80, Vol 1, No 1, 1986.Google Scholar
  32. Mooney RJ, Bennett SW, A Domain Independent Explanation-Based Generalizer, Proceedings of AAAI-86, Fifth National Conference on Artificial Intelligence, Vol I, pp 551–555, Philadelphia, PA, August 1986.Google Scholar
  33. Nordhausen B, Conceptual Clustering Using Relational Information, Proceedings of AAAI-86, Fifth National Conference on Artificial Intelligence, Vol I, pp 508–512, Philadelphia, PA, August 1986.Google Scholar
  34. Pao Yoh-Han, Hu Chi-Heng, Processing of Pattern Based Information, Part I: Inductive Inference Methods Suitable For Use in Pattern Recognition and Artificial Intelligence, Technical Report 101–82,,Case Institute of Technology, Case Western Reserve University, Cleveland, Ohio 44106, 1982.Google Scholar
  35. Quinlan JR, Learning Efficient Classification Procedures and Their Application to Chess End Games, Chapter in Machine Learning: An Artificial Intelligence Approach, Michalski RS, Carbonell JG, Mitchel TM (eds.), Tioga Publishing Co., pp 463–482, 1983.Google Scholar
  36. Quinlan JR, The Effect of Noise on Concept Learning, Chapter in Machine Learning: An Artificial Intelligence Approach, Vol II, Michalski RS, Carbonell JG, Mitchell TM (eds.), Morgan Kaufmann Publishers, pp 149–166, 1986.Google Scholar
  37. Reinke R, Knowledge Acquisition and Refinement Tools for the ADVISE Meta-Expert System, Thesis MS, Department of Computer Science, University of Illinois, Urbana, Illinois 1984.Google Scholar
  38. Rendell LA, A New Basis For State-Space Learning Systems and a Successful Implementation, Artificial Intelligence, Vol 20, No 4, pp 369–392, 1983CrossRefGoogle Scholar
  39. Rose D, Langley P, STAHLp: Belief Revision in Scietific Discovery, AAAI-86, Fifth National Conference on Artificial Intelligence, Vol I, pp 528–532, Philadelphia, PA, August 1986.Google Scholar
  40. Stepp RE, Conjunctive Conceptual Clustering: A Methodology and Experimentation, PhD Thesis, Department of Computer Science, University of Illinois, Urbana, 1984.Google Scholar
  41. Stepp RE, Michalski RS, Conceptual Clustering: Inventing Goal-Oriented Classification of Structured Objects, Machine Learning: An Artificial Intelligence Approach Vol II, Michalski RS, Mitchell TM, Carbonell JG (eds.), Morgan Kaufmann, 1986.Google Scholar
  42. Waltz D, Genesereth M, Hart P, Hendrix G, Joshi A, McDermott J, Mitchel T, Nilsson N, Wilensky R, Artificial Intelligence: An Assessment of the State-of-the-Art and Recommendation for the Future Directions, Al Magazine, Vol 4, No 3, Fall 1983.Google Scholar
  43. Winston PH, Learning Structural Descriptions From Examples, Tech. Report Al TR-213, MIT, AI Lab, Cambridge, MA, 1977.Google Scholar
  44. Winston PH, Learning by Augmenting Rules and Accumulating Sensors, Chapter in Machine Learning: An Artificial Intelligence Approach Vol II, Michalski RS, Mitchell TM, Carbonell JG (eds.), Morgan Kaufmann Publishers, Inc. 1986.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1987

Authors and Affiliations

  • Ryszard S. Michalski
    • 1
  1. 1.Department of Computer ScienceUniversity of IllinoisUrbanaUSA

Personalised recommendations