SAT-based algorithms for inconsistency-tolerant query answering with priorities
This paper presents the first SAT encodings for Pareto- and completion-optimal repairs under general priority relations in prioritized knowledge bases. The authors investigate practical algorithms for inconsistency-tolerant query answering under three semantics (AR, IAR, and brave) based on two notions of optimal repairs. They propose several ways to employ existing and new encodings to compute answers using different reasoning modes of SAT solvers. The work addresses the (co)NP-complete data complexity for a large class of logical theories, extending prior SAT-based procedures that only handled no priority or special structures.
Key facts
- The paper introduces the first SAT encodings for Pareto- and completion-optimal repairs with general priority relations.
- Three semantics are considered: AR, IAR, and brave.
- Two notions of optimal repairs are used: Pareto and completion.
- Deciding query answers under these semantics is (co)NP-complete in data complexity for a large class of logical theories.
- Prior SAT-based procedures only existed for no priority or special priority structures.
- The paper proposes multiple ways to use encodings with different SAT solver reasoning modes.
- The work is published on arXiv with ID 2202.07980.
- The announcement type is replace-cross.
Entities
Institutions
- arXiv