Skip to main navigation Skip to search Skip to main content

Anti-unification for unranked terms and hedges

  • Johannes Kepler University Linz
  • CSIC - Research Institute of Artificial Intelligence

Research output: Contribution to journalScientific articlepeer-review

25 Citations (Scopus)

Abstract

We study anti-unification for unranked terms and hedges that may contain term and hedge variables. The anti-unification problem of two hedges -1 and -2 is concerned with finding their generalization, a hedge q such that both -1 and -2 are instances of q under some substitutions. Hedge variables help to fill in gaps in generalizations, while term variables abstract single (sub)terms with different top function symbols. First, we design a complete and minimal algorithm to compute least general generalizations. Then, we improve the efficiency of the algorithm by restricting possible alternatives permitted in the generalizations. The restrictions are imposed with the help of a rigidity function, which is a parameter in the improved algorithm and selects certain common subsequences from the hedges to be generalized. The obtained rigid anti-unification algorithm is further made more precise by permitting combination of hedge and term variables in generalizations. Finally, we indicate a possible application of the algorithm in software engineering.

Original languageEnglish
Pages (from-to)155-190
Number of pages36
JournalJournal of Automated Reasoning
Volume52
Issue number2
DOIs
Publication statusPublished - Feb 2014

Keywords

  • Anti-unification
  • Generalization
  • Hedges
  • Software clones
  • Unranked terms

Fingerprint

Dive into the research topics of 'Anti-unification for unranked terms and hedges'. Together they form a unique fingerprint.

Cite this