- Andreas Herzig (CNRS and IRIT, France), joint invited speaker with NMR 2020
- Magdalena Ortiz (TU Wien, Austria)
- Balder ten Cate (Google LLC, USA)
Knowledge Base Repair: From Active Integrity Constraints to Active TBoxes
In the database literature it has been proposed to resort to active integrity constraints in order to restore database integrity. Such active integrity constraints consist of a classical integrity constraint together with a set of preferred update actions that can be triggered when the constraint is violated. In this overview we start by reviewing the main repairing routes that have been proposed in the literature. We do so from the perspective of Dynamic Logic, viewing active integrity constraints as programs that test whether a constraint is violated and if so perform appropriate update actions. We then discuss how these ideas can be adapted to Description Logics. We assume extensions of TBox axioms by update actions that denote the preferred ways an ABox should be repaired in case of inconsistency with the axioms of the TBox. The extension of the TBox axioms with these update actions constitute new, active TBoxes.
Between the Open- and Closed-World Assumptions: Querying and Rewriting under Partial Completeness
The semantics of Description Logic (DL) ontologies adopts the open-world assumption (OWA) from classical first-order logic (FOL), which makes them well suited to reason about incomplete knowledge. However, there are many applications where incomplete and complete data co-exist, and the standard OWA, which assumes that all data is incomplete, allows to draw less inferences than desired. Adding closed-predicates to ontology-mediated queries (OMQs) is an elegant and powerful way to reach better conclusions leveraging partial completeness, but unfortunately, it is well-known that OMQs with closed predicates are computationally costly. Their query answering problem is coNP-hard in practically every non-trivial setting, and hence they are not FOL-rewritable. In this talk, we discuss some advantages and challenges of OMQs with closed predicates, and recall a selection of recent results on their rewritability into FOL and into fragments of Datalog.
Balder ten Cate
The Homomorphism Lattice, Unique Characterizations, and Concept Learning
Various problems arising in the context of example-driven approaches to query discovery have turned out to be intimately related to basic structural properties of the homomorphism lattice of finite structures, such as density, or the existence of duals. In this keynote, I will review some such connections, and highlight some relevant recent results.