Learning nonrecursive definitions of relations with linus

  • Nada Lavrač
  • Sašo Džeroski
  • Marko Grobelnik
Part 4: Theorem Proving And EBL
Part of the Lecture Notes in Computer Science book series (LNCS, volume 482)


Many successful inductive learning systems use a propositional attribute-value language to represent both training examples and induced hypotheses. Recent developments are concerned with systems that induce concept descriptions in first-order logic. The deductive hierarchical database (DHDB) formalism is a restricted form of Horn clause logic in which nonrecursive logical definitions of relations can be expressed. Having variables, compound terms and predicates, the DHDB formalism allows for more compact descriptions of concepts than an attribute-value language. Our inductive learning system LINUS uses the DHDB formalism to represent concepts as definitions of relations. The paper gives a description of LINUS and presents the results of its successful application to several inductive learning tasks taken from the machine learning literature. A comparison with the results of other first-order learning systems is given as well.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Bratko, I., Mozetič, I. & Lavrač, N. (1989) KARDIO: A study in deep and qualitative knowledge for expert systems. Boston, MA: The MIT Press.Google Scholar
  2. Cestnik, B. & Bratko, I. (1988) Learning redundant rules in noisy domains. Proc. European Conference on Artificial Intelligence, ECAI-88, Muenchen, Germany.Google Scholar
  3. Cestnik, B., Kononenko, I. & Bratko, I. (1987) ASSISTANT 86: A knowledge-elicitation tool for sophisticated users. In: Bratko, I. & Lavrač, N. (eds.) Progress in machine learning. Wilmslow: Sigma Press.Google Scholar
  4. Clark, P. & Niblett, T. (1989) The CN2 induction algorithm. Machine Learning 1 (3), 261–284. Kluwer Academic Publishers.Google Scholar
  5. Dietterich, T.G. & Michalski, R.S. (1986) Learning to predict sequences. In: Michalski, R.S., Carbonell, J.G. & Mitchell, T.M. (eds.) Machine learning: An artificial intelligence approach (Volume 2). Los Altos: Morgan Kaufmann.Google Scholar
  6. Gams, M. (1989) New measurements highlight the importance of redundant knowledge. Proc. European Working Session on Learning, EWSL 89. Montpellier, France: Pitman.Google Scholar
  7. Lavrač, N. (1990) Principles of knowledge acquisition in expert systems. PhD Thesis, Maribor University, Yugoslavia.Google Scholar
  8. Lavrač, N., Džeroski, S. & Grobelnik, M. (1990) Experiments in learning nonrecursive definitions of relations with LINUS. Report IJS-DP-5863, Jožef Stefan Institute, Ljubljana, Yugoslavia.Google Scholar
  9. Lavrač, N. & Mozetič, I. (1988) Experiments with inductive learning programs NEW-GEM and ASSISTANT in the DHDB environment. Report IJS-DP-5029, Jožef Stefan Institute, Ljubljana, Yugoslavia.Google Scholar
  10. Lloyd, J.W. (1987) Foundations of logic programming (Second edition). Springer-Verlag.Google Scholar
  11. Michalski, R.S., Mozetič, I., Hong, J. & Lavrač, N. (1986) The multi-purpose incremental learning system AQ15 and its testing application on three medical domains. Proc. National Conference on Artificial Intelligence, AAAI-86. Philadelphia, PA: Morgan Kaufmann.Google Scholar
  12. Mozetič, I. (1985) NEWGEM: Program for learning from examples — Technical documentation and user's guide. Report, University of Illinois at Urbana-Champaign, Department of Computer Science. Also: Report IJS-DP-4390, Jožef Stefan Institute, Ljubljana, Yugoslavia.Google Scholar
  13. Mozetič, I. (1987) Learning of qualitative models. In: Bratko, I. & Lavrač, N. (eds.) Progress in machine learning. Wilmslow: Sigma Press.Google Scholar
  14. Mozetič, I. & Lavrač, N. (1988) Incremental learning from examples in a logic-based formalism. Proc. Int. Workshop on Machine Learning, Meta-reasoning and Logic, Sesimbra, Portugal.Google Scholar
  15. Muggleton, S.H. & Buntine, W. (1988) Machine invention of first-order predicates by inverting resolution. Proc. Machine Learning Conference, Ann Arbor, Michigan.Google Scholar
  16. Muggleton, S.H., Bain, M., Hayes-Michie, J. & Michie, D. (1989) An experimental comparison of human and machine learning formalisms. Proc. Sixth International Workshop on Machine Learning. Ithaca, NY: Morgan Kaufmann.Google Scholar
  17. Muggleton, S.H. & Feng, C. (1990) Efficient induction of logic programs. Proc. First Conference on Algorithmic Learning Theory. Tokyo: Ohmsha.Google Scholar
  18. Quinlan, J.R. (1986) Induction of decision trees. Machine Learning 1 (1), 81–106. Kluwer Academic Publishers.Google Scholar
  19. Quinlan, J.R. (1987) Generating production rules from decision trees. Proc. Int. Joint Conference on Artificial Intelligence, IJCAI 87. Milano, Italy: Morgan Kaufman.Google Scholar
  20. Quinlan, J.R. (1989) Learning relations: Comparison of a symbolic and a connectionist approach. Technical Report 346, Basser Dept. Comp. Sc., University of Sydney, Sydney, Australia.Google Scholar
  21. Quinlan, J.R. (1990) Learning logical definitions from relations. To appear in Machine Learning. Kluwer Academic Publishers.Google Scholar
  22. Winston, P.H. (1975) Learning structural descriptions from examples. In: Winston, P.H. (ed.) The psychology of computer vision. New York: McGraw-Hill.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1991

Authors and Affiliations

  • Nada Lavrač
    • 1
  • Sašo Džeroski
    • 1
  • Marko Grobelnik
    • 1
  1. 1.Jožef Stefan InstituteLjubljanaYugoslavia

Personalised recommendations