uml– linked data - inspire.ec.europa.eu · rules for mapping from the source uml geo metamodel to...
TRANSCRIPT
UML– Linked DataDutch best practice for transforming UML Application Schema to Linked
Data Ontologies
The approach centred around developing transformation rules for mapping from the source UML geo metamodel to the
target metamodel(s) for SKOS/RDF/SHACL/OWL implementations. The rules were developed on the basis of the existing metamodels but also on a use case about a test
case domain model: the Golf Course Domain Model and a related dataset
Paul [email protected] – Antwerp 2018
Content
UML legacy
Linked data revelation
UML – LD combination
UML – LD derivation
UniverseUniverse of Discourse(UoD)
model of model of
RDFmodel
UMLmodel transformation
Principal goal
UML legacy
NEN 3610 is de standaard voor het uitwisselen van geo-informatie, gebruikt UML als formele taal voor het vastleggen van semantiek en beveelt GML aan als technische uitwisselingsformat. NEN 3610 is hiermee niet voorbereid om semantiek en gegevensdeling en uitwisseling middels linked data te realiseren.
ISO TC/211
National standards
Domain standards
NEN3610Dutch GeoBaseModel
IMOOV
IMKL
IMKA
D
IMNAB
IMLG
IMWA
IMRO
IMKICH
IMWE
IMGeo
IM01
01IM
BRO
IMTO
PIM
Metingen
IM . . .
ISOISO
INSPIRE European practice
IMAE
R
Data exchange
Pyramid of specialzation
Silos extending common rulesStack of profiles and extensions
Pyramid of reuse and reference
reuse and references of ontologies
Universe of discourse for everyone
Anybody can say anything about anything
Use case defines universe of discourse
Domain defines vocabulary – domain standard
UML-GeoBaseModel
(NEN 3610)
Linked data
Functional layer
Centralized vocabulary
Domain specifications
Generic functionalontologies
All concepts
domain-ontologie
Genericontologies time, geometry, location
temporal, identification
tree, road, buliding etc
Subset andreuse of concepts
ISO
Rules and semanticsnational
Domain models
Rules and semanticsinternational
Linked data revelation
Opening up the data Silos
Centralized vocabulary
- NOT IN COMPETITION -Fundamental differences relate to different use
casesUML and LD are complementary
UML model use case: data exchange, harmonized, controlled (environment), handled byapplications, (data quality centred). No surprises!
LD: data publication, open environment, ‘handled by the web’ (data access centred). Be surprised!?
Typical UML domainmodel use cases/concepts
1 codelists2 vocabularies of classes ‐ taxonomy3 data publishing ‐ exchange4 data validation5 vocabularies related, harmonisation
Typical LD model use cases/concepts
1 vocabulary, thesaurus2a Knowledge model (ontology) simple2b Knowledge model (ontology) complex3 web of data (publishing, sharing), semantic web4 Knowledge deduction5 artificial intelligence
UML – LD combinationUML – LD derivation
Let’s make UML and LD interoperable
UniverseUniverse of Discourse(UoD)
model of model of
RDFmodel
UMLmodel transformation
Building on existing knowledge:
ISO 19150-2: 2015 - Geographic information -- Ontology -- Part 2: Rules for developing ontologies in the Web Ontology Language (OWL)
INSPIRE - Guidelines for the RDF encoding of spatial data (ARE3NA)
UniverseUniverse of Discourse(UoD)
model of model of
UMLmodel
autom.transformation manual refinement
UML NEN3610metamodel
RDFGeoBase
metamodel
RDFmodel#2
RDFmodel#1
conforms to conforms to conforms to
• Use standardized vocabularies• Complex datatypes as concepts• Codelist values as concepts• Relating external concepts• Multiple inheritance• UML:Class not same as OWL:Class
RDFGeoBase
metamodel
UniverseUniverse of Discourse(UoD)
model of model of
autom. transformation manual refinementRDFmodel#2
RDFmodel#1
XSDdata
structuur
XMLdata
RDFdata
derived from
conforms to conforms to
Automatedconversioninputinput outputoutput
conforms to conforms to
Input Input
UMLmodel
Relating MetamodelsOMG – 4 layermetamodel
M3
M0
M1
M2
Metamodels compared and mappedUML profile – RDF/SKOS/OWL/SHACL
OWLSHACL
SKOS
Testing on the golfcourse
http://commons.wikimedia.org/wiki/File:Golf_field.svg
1 teeing ground2 water hazard3 Rough4 out of bounds5 sand bunker6 water hazard7 fairway8 green9 Flagstick10 hole
COINS - ontology
OroX - ontology
LD – General ontology
RDFGeoBase
metamodel
UniverseUniverse of Discourse(UoD)
model of model of
autom. transformation manual refinementRDFmodel#2
RDFmodel#1
XSDdata
structuur
XMLdata
RDFdata
derived from
conforms to
conforms to
Automatedconversioninputinput outputoutput
conforms to conforms to
Input Input
UMLmodel
Still work in progress
Conclusions
UML-community and Linked data community are sharing views
UML -> LD conversions are being developed and understood
UML and LD are becoming interoperable