Advertisement

Learning to relate terms in a multiple agent environment

  • P. Brazdil
  • S. Muggleton
Part 7: Multi-agents
Part of the Lecture Notes in Computer Science book series (LNCS, volume 482)

Abstract

In the first part of the paper we describe how different agents can arrive at different (but overlapping) views of reality. Although the agents can cooperate when answering queries, it is often desirable to construct an integrated theory that explains ‘best’ a given reality. The technique of knowledge integration based on an earlier work is briefly reviewed and some shortcomings of this technique are pointed out. One of the assumptions underlying the earlier work was that all agents must use the same predicate vocabulary. Here we are concerned with the problems that can arise if this assumption does not hold. We also show how these problems can be overcome. It is shown that standard machine learning techniques can be used to acquire the meaning of other agent's concepts. The experiments described in this paper employ INTEG.3, a knowledge integration system, and GOLEM, an inductive system based on relative least general generalization.

Keywords

knowledge integration language differences learning concept definitions learning unknown concepts predicate vocabulary learning in distributed systems 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Boose J., Bradshaw J., Kitto C. and Sherma D. (1989): "From ETS to Acquinas: Six Years of Knowledge Acquisition Tool Development," in Proceedings of Third European Workshop on Knowledge Acquisition for Knowledge-Based Systems, J. Boose, B. Gaines and J.G. Ganascia (eds.), Paris, July 1989.Google Scholar
  2. Brazdil P. and Torgo L. (1990): "Knowledge Acquisition via Knowledge Integration", in Current Trends in Artificial Intelligence, B. Wielinga et al. (eds.), IOS Press, Amsterdam, 1990.Google Scholar
  3. Durfee E., Lesser V.R. and Corkill D.D. (1989): "Cooperative Distributed Problem Solving", in The Handbook of Artificial Intelligence, Volume IV, Barr A., Cohen P.R. and Feigenbaum E.A. (eds.), Addison Wesley, 1989.Google Scholar
  4. Gams M. (1989): "The Measurement Highlight the Importance of Redundant Knowledge", in Proceedings of 4th European Working Session on Machine Learning (EWSL-89), K. Morik (ed.), pp. 71–80, Pitman — Morgan Kaufmann.Google Scholar
  5. Muggleton S. and Feng C. (1990): "Efficient Induction of Logic Programs", in Proceedings of the First Conference on Algorithmic Learning, Tokyo, Japan, October 1990, Ohmsha Publ., Tokyo.Google Scholar
  6. Murray K.S. and Porter B.W. (1989): "Controlling Search for the Consequences of New Information During Knowledge Integration", in Proceedings of 6th International Workshop on Machine Learning, A.M. Segre (ed.), Ithaca, New York, Morgan Kaufmann Inc.Google Scholar
  7. Shaw M. and Gaines B. (1989): Knowledge Acquisition: Some Foundations, manual Methods and Future Trends, in Proceedings of Third European Workshop on Knowledge Acquisition for Knowledge-Based Systems, J. Boose, B. Gaines and J.G. Ganascia (eds.), Paris, July 1989.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1991

Authors and Affiliations

  • P. Brazdil
    • 1
  • S. Muggleton
    • 2
  1. 1.LIACC-CIUPPortoPortugal
  2. 2.The Turing InstituteGlasgowUK

Personalised recommendations