Single Elementary Associative-Commutative Matching

  • Authors:
  • Steven Eker

  • Affiliations:
  • Computer Science Laboratory, SRI International, Menlo Park, CA 94025, USA. e-mail: eker@csl.sri.com

  • Venue:
  • Journal of Automated Reasoning
  • Year:
  • 2002

Quantified Score

Hi-index 0.00

Visualization

Abstract

A single elementary associative-commutative (AC) matching problem has a pattern term that consists of a single (variadic) AC function symbol with only variable symbols as arguments and a subject term that consists of a single (variadic) AC function symbol with only constant symbols as arguments. We show that even this very restricted formulation of AC matching has an NP-complete decision problem. We consider a number of methods to contain the growth in the search space, including a lookup table for the solubility of subproblems, a digraph reformulation of the problem, and a search tree pruning method that uses failure information together with a partial ordering on branches. We give empirical results for the method that seems to work best in practice, and we list some ‘hard’ problem instances.