This group aims to point at research on Linked Data and to identify relations to and impacts on INSPIRE.

| Share
Group discussion > Semantic Web at the Aalborg conference

Semantic Web at the Aalborg conference

Paul van Genuchten
1094 days ago

At the 2014 edition of the INSPIRE conference in Aalborg many presentations referenced semantic concepts. Tracks were organised on vocabularies, linked data and registers. Semantic concepts were also mentioned in others tracks like: 'data-discovery' and 'Open data'. An important reason for this amount of interest lies in the fact that the linked data community is a big potential audience for INSPIRE data. Other reasons include the fact that semantic technology can provide functionality that is not easily available in the traditional OGC W*S standards, like querying an SDI on all features around a specific location (in a multitude of datasets). A third and potentially underestimated reason is the fact that semantic technology can facilitate the process of data harmonisation that Annex 1 data providers are confronted with in the upcoming two years. I'd like to guide you through some of the presentations on that conference I happened to attend and thought were interesting in this area. An interesting presentation by Monica de Martino rom the eEnvPlus team (http://inspire.ec.europa.eu/events/conferences/inspire_2014/pdfs/19.06_5_11.00_Monica_De_Martino.pdf), they did research on vocabularies that are relevant within the environment scope. They evaluated each of the relevant vocabularies on some quality indicators like the license, multi-linguality and popularity. Also they evaluated some software to match concepts in different vocabularies, a basic prerequisite to make the semantic web work. Another relevant development is the upcoming registry software products implementing iso19135. Most prominent the INSPIRE registry itself. Similar initiatives are available in germany (http://inspire.ec.europa.eu/events/conferences/inspire_2014/pdfs/19.06_5_11.00_Marcus_Walther.pdf) and in GeoNetwork (https://trac.osgeo.org/geonetwork/wiki/registers). The INSPIRE registry (soon offering a SKOS RDF/XML encoding on the included concepts) manages all the concepts used within the INSPIRE domain, including concepts from the regulation documents, schema's, codelists and soon hopefully the technical guidelines (http://inspire.ec.europa.eu/registry). To be researched is if registers can be relevant in maintaing cross-vocabulary links. Another interesting presentation slightly related was by François Prunayre and Vincent Bombaerts (http://inspire.ec.europa.eu/events/conferences/inspire_2014/pdfs/18.06_6_16.00_Francois_Prunayre.pdf). They gave a quick overview of the recently relased iso19115-1/3/4 metadata standards and how they are implemented for Belgium in Geonetwork. The new standards provide a number of improvements, some of these better facilitate the semantic and open data world. For example the integration of iso19110 feature catalog metadata (to facilitate providing semantics for dataset attributes), usage of codelists for defining license restrictions, and the notion of a 'point-of-truth' url, which can potentially be used as a PID. On friday Paul Jansen concluded the set of semantics tracks with a presentation on a participative approach to manage the alignment of ontologies. A very practical use case that shows that one needs to align ontologies before you can start the harmonisation of datasets. Alignment of ontologies can best be managed in a participative approach, including stakeholders of the practical and modelling world (http://inspire.ec.europa.eu/events/conferences/inspire_2014/pdfs/20.06_4_09.00_Paul_Janssen.pdf).

As GeoCat we're a participant in the Citizen Observatories WEB FP7 Project (http://cobwebproject.eu), lead by University of Edinburgh. One of the main goals of this project is the conflation and validation of Citizen Science Observations in Unesco Biospheres with authoritive datasets. As the Biospheres involved are located in Whales, Greece and Germany. The INSPIRE datamodels are quite relevant in providing a common base for the authoritive data used. At this point in time our assumption is that semantic technologies best meet our requierements to manage about data conflation. It explains our interest in these matters, although we also get frequent queries on the amount of RDF support currently available in GeoNetwork (which is considerable). I hope above provides a good overview of the current state of semantic technologies in the scope of INSPIRE. Feel free to hook up your initiative in this thread.

Permalink