Learning recursion with iterative bootstrap induction (Extended abstract)

  • Alípio Jorge
  • Pavel Brazdil
Extended Abstracts
Part of the Lecture Notes in Computer Science book series (LNCS, volume 912)


In this paper we are concerned with the problem of inducing recursive Horn clauses from small sets of training examples. The method of iterative bootstrap induction is presented. In the first step, the system generates simple clauses, which can be regarded as properties of the required definition. Properties represent generalizations of the positive examples, simulating the effect of having larger number of examples. Properties are used subsequently to induce the required recursive definitions. This paper describes the method together with a series of experiments. The results support the thesis that iterative bootstrap induction is indeed an effective technique that could be of general use in ILP.


  1. [1]
    Aha D W, Lapointe S, Ling C X, Matwin S (1994): “Inverting Implication with Small Training Sets”, in Proceedings of the European Conference on Machine Learning, ECML-94, ed. F. Bergadano and L. de Raedt, Springer Verlag.Google Scholar
  2. [2]
    Brazdil P, Jorge A. (1994): “Learning by Refining Algorithm Sketches”, in Proceedings of ECAI-94, T. Cohn (ed.), Amsterdam, The Netherlands.Google Scholar
  3. [3]
    Cohen W W (1993): “Rapid prototyping of ILP systems using explicit bias” in Proceedings of 1993 IJCAI Workshop on ILP.Google Scholar
  4. [4]
    Idestam-Almquist P (1993) “Generalization under implication by recursive antiunification”, in Proceedings of ILP-93, Jozef Stefan Institute, technical report.Google Scholar
  5. [5]
    Michalski R S, (1994): “Inferential Theory of Learning: Developing Foundations for Multistrategy Learning”, in Machine Learning, A Multistrategy Approach, Volume IV, Ryszard Michalski and Gheorghe Tecuci, Morgan Kaufmann.Google Scholar
  6. [6]
    Muggleton S. (1993): “Inductive Logic Programming: derivations, successes and shortcomings” in Proceedings of ECML-93, P. Brazdil (ed.), Springer-Verlag.Google Scholar
  7. [7]
    Zelle J M, Mooney R J, Konvisser J B, (1994):“Combining Top-down and Bottom-up Techniques in Inductive Logic Programming” in Proceedings of the Eleventh International Conference on Machine Learning ML-94, Morgan-Kaufmann.Google Scholar
  8. [8]
    Richards B, Mooney R (1992): “Learning relations by pathfinding” in Proceedings of the Tenth National Conference on Artificial Intelligence, Cambridge, MA, MIT Press.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1995

Authors and Affiliations

  • Alípio Jorge
    • 1
  • Pavel Brazdil
    • 1
  1. 1.LIACCUniversity of PortoPortoPortugal

Personalised recommendations