Semantic Generalizations for Proving and Disproving Conjectures by Analogy

  • Authors:
  • Gilles Défourneaux;Christophe Bourely;Nicolas Peltier

  • Affiliations:
  • LEIBNIZ -- IMAG, 46, av. Félix Viallet, 38 031 Grenoble Cedex, France;LEIBNIZ – IMAG, 46, av. Félix Viallet, 38 031 Grenoble Cedex, France;LEIBNIZ – IMAG, 46, av. Félix Viallet, 38 031 Grenoble Cedex, France

  • Venue:
  • Journal of Automated Reasoning
  • Year:
  • 1998

Quantified Score

Hi-index 0.00

Visualization

Abstract

Taking an extension of resolution as a base calculus (though the sameprinciples are applicable to other calculi) for searching proofs(refutations) and counterexamples (models), we introduce a new method ableto find refutations and also models by analogy with refutations and modelsin a knowledge base. The source objects for the analogy process aregeneralizations of the refutations (models). They are included in theknowledge base, and then higher-order matching techniques for the choice ofthe relevant source objects as well as the building of a new proof or amodel by analogy are used. Some comparisons with existing methods as well astwo detailed running examples on generalization show evidence of theinterest of our approach.