Advertisement

A machine learning tool designed for a model-based knowledge acquisition approach

  • Jérôme Thomas
  • Philippe Laublet
  • Jean-Gabriel Ganascia
Problem Solving Models Support Tools
Part of the Lecture Notes in Computer Science book series (LNCS, volume 723)

Abstract

In this paper, we present a new system, ENIGME. Its purpose is to learn the operative domain knowledge in the form of a rule system that follows as closely as possible the explicit reasoning method chosen for the future system. To this end we have mapped the inputs of a Machine Learning system to different parts of the model of expertise as it is used in the KADS methodology. This knowledge constrains the learning process thereby assuring coherence between the learnt and acquired knowledge, and allowing the expert to guide the learning tool effectively.

Keywords

Empirical learning knowledge acquisition conceptual modelling 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. [Buntine 90]
    W. Buntine and D. Stirling, Interactive induction, in Machine Intelligence 12, Oxford University Press.Google Scholar
  2. [Clancey 85]
    William J. Clancey, Heuristic Classification, University of Stanford report STAN-CS-85-1066, 1985Google Scholar
  3. [Chandrasekaran 93]
    B. Chandrasekaran and T.R. Johnson, Generic Tasks and Task Structures: History, Critique and new Directions. In J-M. David et al. (eds.) Second generation of expert systems (Springer Verlag).Google Scholar
  4. [Eshelman 88]
    Larry Eshelman, Mole: a Knowledge acquisition tool for cover and differentiate systems, in Automating KA for expert systems, S. Marcus (ed.), Kluwer Academic PublishersGoogle Scholar
  5. [Ganascia 87]
    Jean-Gabriel Ganascia, AGAPE et CHARADE: deux techniques d'apprentissage symbolique appliquées à la construction de bases de connaissances. Thèse de doctorat d'état de Université de Paris-Sud (Orsay).Google Scholar
  6. [Ganascia 88]
    Jean-Gabriel Ganascia, Improvement and refinement of the learning bias semantic, in proceedings of ECAI, Y. Kodratof (ed.), Munich, 1988.Google Scholar
  7. [Laublet 92]
    P. Laublet, T. Scutt, J. Thomas, K. O'Hara, N. Major and N. Shadbolt, Integration of Machine Learning Techniques and Tools in VITAL, VITAL internal deliverable ONERA/T2.1.4/I/1Google Scholar
  8. [Le Roux 93]
    Bernard Le Roux and Philippe Laublet, From simple diagnosis to cover and differentiate: a trial for a conceptual and compositional approach, in proceedings of the third KADS Meeting (Siemens AG, Munich)Google Scholar
  9. [McDermott 88]
    John McDermott, Preliminary steps toward a taxonomy of problem solving methods. In Automating KA for expert systems, S. Marcus (ed.), Kluwer Academic PublishersGoogle Scholar
  10. [Mitchell 80]
    Tom M. Mitchell, The Need for Biases in Learning Generalizations, In Readings in Machine Learning, J. W. Shavlik and T. G. Dietterich (eds.), Morgan KaufmannGoogle Scholar
  11. [Morik 91]
    Katharina Morik and Edgar Sommer, Balanced Cooperative Modelling using Mobal 1.0: user guide. MLT ESPRIT project deliverable GMD/P2154/21/1Google Scholar
  12. [O'Hara 92]
    K. O'Hara, N. Shadbolt, Philippe Laublet, M. Zacklad and Bernard Le Roux, Knowledge Acquisition Methodology, VITAL deliverable 2.1.2Google Scholar
  13. [Puerta 92]
    A.R. Puerta, J.W. Egar, S.W. Tu and M.A. Musen, A multiple method knowledge-acquisition shell for the automatic generation of knowledge-acquisition tools. Knowledge Acquisition V4, N 2, June 1992Google Scholar
  14. [Russel 90]
    Stuart J. Russel and Benjamin N. Grosof, Declarative Bias: An Overview, in Change of representation and inductive bias, P. Benjamin (ed.)Google Scholar
  15. [Shadbolt 90]
    Nigel Shadbolt and B.J. Wielinga, Knowledge based knowledge acquisition: the next generation tools, in current trends in knowledge acquisition, B. Wielinga et al. (eds.), IOS Press.Google Scholar
  16. [Steels 91]
    Luc Steels, COMMET: A componentional methodology for knowledge engineering. ESPRIT project CONSTRUCT. Deliverable WP 2/3/4 (synthesis).Google Scholar
  17. [Thomas 93]
    Jérôme Thomas, Jean-Gabriel Ganascia et Philippe Laublet, Apprentissage dirigé par des modèles de connaissances, Actes des Journées d'Acquisition, de Validation et d'Apprentissage, St Raphaël 1993.Google Scholar
  18. [van Heijst 92]
    G. van Heijst, P. Terpstra, B. Wielinga and N. Shadbolt, Using Generalised Directive Models in Knowledge Acquisition, in Th. Wetter, K.D. Althoff, J. Boose, B.R. Gaines, M. Linster, and F. Schmalhofer (eds.), Current Developments in Knowledge Acquisition-EKAW'92, Springer Verlag.Google Scholar
  19. [Webb 91]
    Geoffrey I. Webb, Einstein: An interactive inductive knowledge acquisition tool, sixth Banff Knowledge Acquisition for Knowledge Based Systems Workshop, Canada 1991Google Scholar
  20. [Wielinga 92]
    B.J. Wielinga, A.Th. Schreiber and J.A. Breuker, KADS: A Modelling Approach to Knowledge Engineering. Knowledge Acquisition volume 4, number 1, March 1992.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1993

Authors and Affiliations

  • Jérôme Thomas
    • 1
    • 2
    • 3
  • Philippe Laublet
    • 1
  • Jean-Gabriel Ganascia
    • 2
  1. 1.ONERA, DMI, GIAChatillon Cedex
  2. 2.LAFORIA-IBP Université P. et M. CurieParis Cedex
  3. 3.Department of PsychologyAI GroupUniversity Park

Personalised recommendations