We want to device best practices on how to publish the related metadata in order to enable automatic processing. We shall devise best practices and use these in the web pages of our institutions to demonstrate the use of such metadata. A possible minimal core to start with would be to focus on (a taxonomy for) a particular area, like for instance computer science research institutions. Here, we can define properties to link persons, publications, conferences (or more general events), and institutions with certain areas of expertise.
We should strive for
- identifying widely agreed and usable vocabularies and taxonomies to use for expertise annotation, such as e.g. the ACM classifications or maybe also the subject areas at the WWW Conferences Archive could be interesting. RDFizing these categories would already be quite beneficial for annotations in the Computer Science area. The WWW Conferences Archive is connected to the Eprints initiative, which aims at making research freely available. Another possibly related metadata standard is IEEE Learning Object Metadata (IEEE LOM); there already exist some guidelines how to use IEEE LOM in combination with URIs from the [Dublin Core http://dublincore.org/] initiative. Furthermore, there already exist initiatives for creating XML standards for CV's, such as for instance: http://xmlresume.sourceforge.net/, or http://www.tsphethean.co.uk/itp3104/index.html. An example for a general events ontology can be found at http://www.eswc2006.org/technologies/ontology . Addresses of institutions or persons could for example be expressed using vCard. As for the description of research projects, ther DOAP (Description of a project) voacabulary, might be a good starting point.
Several projects in the Semantic Web realm have have already created their own ontologies for describing events, organisations, e.g.
- the KnowledgeWeb platform ontologies: http://knowledgeweb.semanticweb.org/semanticportal/OWL/
- the AKT portal ontology: http://www.aktors.org/ontology/portal
- the SWRC portal ontology: http://swrc.ontoware.org/ontology
- the DERI Semantic Web Portal (SWP) working group's ontology: http://sw-portal.deri.org/ontologies/swportal
However, these ontologies seem to be developed from scratch and (i) neither be based in pre-existing vocabularies or defacto standards, (ii) nor be adopted outside the project where they have been developed, which are two major objective in our initiative. The DERI SW-portal ontology is partly an exception, it tries to reuse at least foaf, rss and bibtex. Best Practice Recipes for Publishing RDF Vocabularies is another document that we must be in our minds.
- several international institutions actively participating in this effort and adopting the results by implementing the devised best practices on their institution web pages to achieve critical mass for getting a consensual, usable core started quickly. The ExpertFinder Initiative is obviously related to community efforts such as the EU's successful CORDIS, and could be positioned as as a semantic enrichment of CORDIS and a refinement to the level of individual researchers, in the direction of decentralization and enrichment of the available information stored in such portals.