Roman Kontchakov

Ontology-Based Data Access: Ontop of Databases

We present the architecture and technologies underpinning the OBDA system Ontop and taking full advantage of storing data in relational databases. We discuss the theoretical foundations of Ontop: the tree-witness query rewriting,

Reasoning over extended ER models

\mathcal{T}

Ontop: Answering SPARQL queries over relational databases

-mappings and optimisations based on database integrity constraints and SQL features. We analyse the performance of Ontop in a series of experiments and demonstrate that, for standard ontologies, queries and data stored in relational databases, Ontop is fast, efficient and produces SQL rewritings of high quality.

The price of query rewriting in ontology-based data access

We investigate the computational complexity of reasoning over various fragments of the Extended Entity-Relationship (EER) language, which includes a number of constructs: ISA between entities and relationships, disjointness and covering of entities and relationships, cardinality constraints for entities in relationships and their refinements as well as multiplicity constraints for attributes. We extend the known EXPTIME-completeness result for UML class diagrams [5] and show that reasoning over EER diagrams with ISA between relationships is EXPTIME-complete even without relationship covering. Surprisingly, reasoning becomes NP-complete when we drop ISA between relationships (while still allowing all types of constraints on entities). If we further omit disjointness and covering over entities, reasoning becomes polynomial. Our lower complexity bound results are proved by direct reductions, while the upper bounds follow from the correspondences with expressive variants of the description logic DL-Lite, which we establish in this paper. These correspondences also show the usefulness of DL-Lite as a language for reasoning over conceptual models and ontologies.

On the computational complexity of decidable fragments of first-order linear temporal logics

We present Ontop, an open-source Ontology-Based Data Access (OBDA) system that allows for querying relational data sources through a conceptual representation of the domain of interest, provided in terms of an ontology, to which the data sources are mapped. Key features of Ontop are its solid theoretical foundations, a virtual approach to OBDA, which avoids materializing triples and is implemented through the query rewriting technique, extensive optimizations exploiting all elements of the OBDA architecture, its compliance to all relevant W3C recommendations (including SPARQL queries, R2RML mappings, and OWL2QL and RDFS ontologies), and its support for all major relational databases.

Exponential lower bounds and separation for query rewriting

We give a solution to the succinctness problem for the size of first-order rewritings of conjunctive queries in ontology-based data access with ontology languages such as OWL 2 QL, linear Datalog± and sticky Datalog±. We show that positive existential and nonrecursive datalog rewritings, which do not use extra non-logical symbols (except for intensional predicates in the case of datalog rewritings), suffer an exponential blowup in the worst case, while first-order rewritings can grow superpolynomially unless NP ⊆ P/poly. We also prove that nonrecursive datalog rewritings are in general exponentially more succinct than positive existential rewritings, while first-order rewritings can be superpolynomially more succinct than positive existential rewritings. On the other hand, we construct polynomial-size positive existential and nonrecursive datalog rewritings under the assumption that any data instance contains two fixed constants.

A Cookbook for Temporal Conceptual Data Modelling with Description Logics

We study the complexity of some fragments of first-order temporal logic over natural numbers time. The one-variable fragment of linear first-order temporal logic even with sole temporal operator /spl square/ is EXPSPACE-complete (this solves an open problem of J. Halpern and M. Vardi (1989)). So are the one-variable, two-variable and monadic monodic fragments with Until and Since. If we add the operators O/sup n/, with n given in binary, the fragment becomes 2EXPSPACE-complete. The packed monodic fragment has the same complexity as its pure first-order part - 2EXPTIME-complete. Over any class of flows of time containing one with an infinite ascending sequence - e.g., rationals and real numbers time, and arbitrary strict linear orders - we obtain EXPSPACE lower bounds (which solves an open problem of M. Reynolds (1997)). Our results continue to hold if we restrict to models with finite first-order domains.

Spatial logics with connectedness predicates

We establish connections between the size of circuits and formulas computing monotone Boolean functions and the size of first-order and nonrecursive Datalog rewritings for conjunctive queries over OWL 2 QL ontologies. We use known lower bounds and separation results from circuit complexity to prove similar results for the size of rewritings that do not use non-signature constants. For example, we show that, in the worst case, positive existential and nonrecursive Datalog rewritings are exponentially longer than the original queries; nonrecursive Datalog rewritings are in general exponentially more succinct than positive existential rewritings; while first-order rewritings can be superpolynomially more succinct than positive existential rewritings.

The Complexity of Clausal Fragments of LTL

We design temporal description logics (TDLs) suitable for reasoning about temporal conceptual data models and investigate their computational complexity. Our formalisms are based on DL-Lite logics with three types of concept inclusions (ranging from atomic concept inclusions and disjointness to the full Booleans), as well as cardinality constraints and role inclusions. The logics are interpreted over the Cartesian products of object domains and the flow of time (ℤ, <), satisfying the constant domain assumption. Concept and role inclusions of the TBox hold at all moments of time (globally), and data assertions of the ABox hold at specified moments of time. To express temporal constraints of conceptual data models, the languages are equipped with flexible and rigid roles, standard future and past temporal operators on concepts, and operators “always” and “sometime” on roles. The most expressive of our TDLs (which can capture lifespan cardinalities and either qualitative or quantitative evolution constraints) turns out to be undecidable. However, by omitting some of the temporal operators on concepts/roles or by restricting the form of concept inclusions, we construct logics whose complexity ranges between NLogSpace and PSpace. These positive results are obtained by reduction to various clausal fragments of propositional temporal logic, which opens a way to employ propositional or first-order temporal provers for reasoning about temporal data models.

Ontology-based data access with databases: a short course

We consider quantifier-free spatial logics, designed for qualitative spatial representation and reasoning in AI, and extend them with the means to represent topological connectedness of regions and restrict the number of their connected components. We investigate the computational complexity of these logics and show that the connectedness constraints can increase complexity from NP to PSpace, ExpTime and, if component counting is allowed, to NExpTime.