MOSIG Master 2ND YEAR Research
YEAR 2011/2012

MASTER TOPIC PROPOSAL

ADVISOR: Jérôme Euzenat

TEL: +33 (0)476 61 53 66

EMAIL: Jerome:Euzenat#inria:fr

TEAM AND LAB: Exmo team, INRIA & LIG

MASTER PROFILE: Artificial intelligence and the web

Reference number: Proposal n°1045

TITLE:

Query generation from expressive ontology alignments

The goal of the semantic web is to take advantage of formalised knowledge (in languages like RDF) at the scale of the worldwide web. In particular, it is based on ontologies which define concepts used for representing knowledge on the web, e.g., for annotating a picture, specifying a web service interface or expressing the relation between two persons. However, it is likely that different information sources and different actors in different contexts will use different ontologies. It is thus necessary to find correspondences between concepts of these ontologies.

A correspondence is defined by the two related entities, which can be classes, instances, properties, formulas as well as combination of those, the relation between these entities (equivalence, subsumption, incompatibility, etc.). A set of such correspondences is called an alignment [1]. Alignments are used for importing data from one ontology to another or for translating queries.

This master topic is related to query generation and translation. It consists of studying query generation and transformation from expressive alignments written in the EDOAL language [2]. The EDOAL language allows for constructing compound entities from the ontology entities. These coumpound entities are then related in a correspondence. For instance, it can express relations like:

An autobiography is a book whose author is the topic;
An INSEE region is a NUTS region of level 2 located in France;
velocity is the same property as speed, except that the former is expressed in miles-per-hour while the latter is in meters per second;

The standard query language for RDF is called SPARQL [3]. It allows for expressing queries corresponding to the EDOAL statements. The first goal of this work is to consider how such queries may be generated and what is missing in the current EDOAL language for doing this properly.

In fact, there are several ways to generate queries from alignments [4]:

query mediator which transform an existing query according to the alignment (so the query will apply to an ontology instead of another, looking for books corresponding to autobiographies)[5];
query generator which generates a query for each related expression in a correspondence (this may be used for retreiving entities from datasets);
data transformation which generates a unique SPARQL query able to transform data from one ontology to another;
CONSTRUCT-based SPARQL link generators able to generate a query for creating links between datasets.

Of course, these modalities are not independent and a second goal of this work is to consider how to articulate them.

Finally, EDOAL expression transformation is far from a simple task because:

we want to rewrite query parts which match the most specific EDOAL expression
it is difficult to recognise them because they may be interleaved with others.

Indeed, it is difficult to predict the interaction between the generated SPARQL queries at runtime. This is a well known problem of term rewritting: that of confluence [6]. We want that whatever the order in which we evaluate the SPARQL queries, this provides the same result. This is trivially true if not two queries can match the same piece of data, not if several queries can concern the same resource.

In the present case, we also would like that all rewriting concerning the same entity be treated within the same query (for both performance and coherence considerations). This problem seems to be well known in database mapping generation and has been adapted for simple correspondences [7].

Expected results

A strategy for generating and transforming SPARQL queries from EDOAL alignments;
A study of the problem raised by optimal set of query generation;
An implementation of query generation/transformation within the Alignment API [8];

References

[1] Jérôme Euzenat, Pavel Shvaiko, Ontology matching, Springer, Berlin (DE), 2007
[2] EDOAL: https://moex.gitlabpages.inria.fr/alignapi/edoal.html
[3] Eric Prud'hommeaux, Andy Seaborne (eds), SPARQL Query Language for RDF, W3C Recommendation, 2008 http://www.w3.org/TR/rdf-sparql-query/
[4] Jérôme Euzenat, François Scharffe, Axel Polleres, Processing ontology alignments with SPARQL (Position paper), in: Proc. IEEE international workshop on Ontology alignment and visualization (OAaV), Barcelona (ES), pp913-917, 2008 https://exmo.inria.fr/files/publications/euzenat2008d.pdf
[5] Gianlucca Correndo, Manuel Salvadores, Ian Millard, Hugh Glaser, Nigel Shadbolt, SPARQL Query rewriting for implementing data integration over linked data. Proc. 1st International Workshop on Data Semantics (DataSem), Lousanne (CH), 2010http://eprints.ecs.soton.ac.uk/18370/
[6] Franz Baader, Tobias Nipkow, Term rewritting and all that, Cambridge university press, Cambridge (UK), 1998 http://www4.in.tum.de/~nipkow/TRaAT/.
[7] Carlos Rivero, Inma Hernández, David Ruiz, Rafael Corchuelo, Generating SPARQL executable mappings to integrate ontologies, Accepted to 30th international conference on conceptual modelling (ER), Brussels (BE), 2011 http://www.tdg-seville.info/Download.ashx?id=224
[8] https://gitlab.inria.fr/moex/alignapi

MOSIG Master 2E ANNÉE Research
ANNÉE 2011/2012

PROPOSITION DE SUJET DE MASTER

RESPONSABLE: Jérôme Euzenat

TÉL: +33 (0)476 61 53 66

ADRESSE ÉLECTRONIQUE: Jerome:Euzenat#inrialpes:fr

LABORATOIRE ET ÉQUIPE: Équipe Exmo, INRIA & LIG

PROFIL DU PROJET: Parcours web et intelligence artificielle

TITRE:

Génération de requêtes à partir d'alignements d'ontologies expressifs

http://exmo.inria.fr/training/M2R-2011-querygen.html

$Id: M2R-2011-querygen.html,v 1.11 2021/12/17 16:04:32 euzenat Exp $