Ontology-based data access (OBDA) is a new paradigm for accessing source databases through mediation of a conceptual domain view, given in terms of an ontology [9]. A major issue in OBDA is the design of an OBDA specification and the management of its evolution. An OBDA specification is constituted by an ontology, usually a Description Logic (DL) TBox, a schema of the source databases, and a declarative mapping specifying the semantic relationship between the data at the sources and the elements of the ontology. In the following we denote it by J = 〈T ,S,M〉, where T is the TBox (a set of DL axioms), S the source schema, i.e., a relational signature possibly with integrity constraints (ICs), and the mappingM is a set of mapping assertions of the form φ(x) ; ψ(x), where φ(x) and ψ(x) are queries over S and T , respectively, both with free variables x (such an assertion is called a GLAV mapping assertion if both φ(x) and ψ(x) are conjunctive queries (CQs), while it is a GAV mapping assertion if it is GLAV and ψ(x) is an atom without occurrences of non-free variables). The design of the specification is normally conducted in an iterative fashion, and changes are continuously implemented to its various components. Also, the entire specification is often a lively artifact, continuously modified due to, e.g., changes in the requirements. Due to these characteristics, OBDA design and maintenance must be supported by adequate tools and methodologies. The mapping is certainly the component of the specification which has received so far less attention, and thus consolidated tools supporting its design are currently not available. Mapping design is a time-consuming and complex operation, which typically (and especially in complex scenarios) has to be conducted manually [1]. Of course, modifying the mapping due to changes in the other components of the specification is tedious and time-consuming as well. In this paper we study the evolution of OBDA specifications. We start our investigation by observing that many approaches exist for both ontology evolution [12] and database schema evolution [11]. However, to the best of our knowledge, no previous study has analyzed evolution in the presence of mappings connecting an ontology to a database schema. In this sense, a problem that is close to OBDA is ontology matching and alignment, which is based on the use of a notion of mapping to integrate different ontologies. Several works have studied the problem of repairing inconsistent mappings in this context (e.g., [3, 7, 8, 10]). However, the framework of ontology matching, and in particular the notion of mapping, is very different from OBDA. We adopt a mapping-centered notion of OBDA evolution: given an OBDA specification J = 〈T ,S,M〉, we want to repair the mappingM given a modification of the
[1]
Guilin Qi,et al.
A Conflict-based Operator for Mapping Revision
,
2009,
Description Logics.
[2]
Ian Horrocks,et al.
Evaluating Mapping Repair Systems with Large Biomedical Ontologies
,
2013,
Description Logics.
[3]
Evgenij Thorstensen,et al.
Towards Mapping Analysis in Ontology-Based Data Access
,
2014,
RR.
[4]
Maurizio Lenzerini,et al.
Developing Ontology-based Data Management for the Italian Public Debt
,
2014,
SEBD.
[5]
Erhard Rahm,et al.
An online bibliography on schema evolution
,
2006,
SGMD.
[6]
Gabriel M. Kuper,et al.
Updating Logical Databases
,
1986,
Adv. Comput. Res..
[7]
Heiner Stuckenschmidt,et al.
Reasoning Support for Mapping Revision
,
2009,
J. Log. Comput..
[8]
Enrico Motta,et al.
Ontology evolution: a process-centric survey
,
2013,
The Knowledge Engineering Review.
[9]
Diego Calvanese,et al.
Linking Data to Ontologies
,
2008,
J. Data Semant..
[10]
Özgür L. Özçep.
Towards Principles for Ontology Integration
,
2008,
FOIS.
[11]
Evgenij Thorstensen,et al.
Mapping Analysis in Ontology-based Data Access: Algorithms and Complexity (Extended Abstract)
,
2015,
Description Logics.