On Computing Maximal Subsets of Clauses that Must Be Satisfiable with Possibly Mutually-Contradictory Assumptive Contexts

Philippe Besnard (1, 2) , Éric Grégoire (3) , Jean-Marie Lagniez (3)

1 IRIT-LILaC - Logique, Interaction, Langue et Calcul
2 CNRS - Centre National de la Recherche Scientifique
3 CRIL - Centre de Recherche en Informatique de Lens

An original method for the extraction of one maximal subset of a set of Boolean clauses that must be satisfiable with possibly mutually contradictory assumptive contexts is motivated and experimented. Noticeably, it performs a direct computation and avoids the enumeration of all subsets that are satisfiable with at least one of the contexts. The method applies for subsets that are maximal with respect to inclusion or cardinality.

Keywords: Partial-max-SAT; SAT; MSS; Maximal Satisfiable Subset

Domains: Computer Science [cs]; Artificial Intelligence [cs.AI]

Philippe Besnard, Éric Grégoire, Jean-Marie Lagniez. On Computing Maximal Subsets of Clauses that Must Be Satisfiable with Possibly Mutually-Contradictory Assumptive Contexts. 29th AAAI Conference on Artificial Intelligence (AAAI'15), 2015, Austin, Texas, United States. pp.3710-3716, ⟨10.1609/aaai.v29i1.9751⟩. ⟨hal-03300455⟩