[foaf-dev] Event model - eon

Veronique Malaise <vmalaise-vHs5IaWfoDhmR6Xm/[email protected]>
Newsgroups gmane.comp.web.rdfweb
Message-ID <[email protected]>
Hi Toby,

Danbri forwarded me the thread below, as he knows that I am working on/ 
interested in Event modeling, along with Willem Robert van Hage, at  
the VU University of Amsterdam.

Context: your thread:
"Let's do a bit of collaborative work to flesh out a FOAF-like
lightweight vocabulary for describing events, suitable for embedding in
RDFa to describe concerts, conferences, birthdays, anniversaries,
national holidays and so on.

Sure, we have the iCalendar RDF vocab, but I've come to the conclusion
that that's more suited to representing calendar entries - not the
events themselves.

I'm thinking of a simple set of extensions for Yves Raimond's Event
Ontology.

...

I've placed a sketch on my data wiki <http://wiki.ontologi.es/>. Feel
free to edit; and add yourself as a foaf:maker if you do!"

Willem and I have developed SEM, an event ontology that indeed extends  
on Yves' one, and maps to it, as well as to LODE, from Raphaël Troncy  
and Ryan Shaw. The ontology is available at:
http://semanticweb.cs.vu.nl/2009/11/sem/
and some papers describing it can be found at:
http://www.few.vu.nl/~wrvhage/
on the Publication tab, yes :)

In two words, the principles behind SEM are
- a very limited set of classes (Core classes) and properties to  
express what, who, where, when and their context: for example the  
roles and authorities that are related to the statements triplified/ 
represented (Constraint classes)
- all the "components" of an event are optional and duplicable, to be  
able to deal with things actually published on the Web. For example,  
you might get 4 different birthdates for Rembrandt, and although he  
was born only one, you're interested in gathering the four values for  
the birthdate rather than not having a single one... or having to  
chose upfront (so independently from a given context)
- as few constraints as possible, to enable as much domain knowledge  
reuse as possible
- no subclasses of events but hierarchies of types for the events,  
actors, places etc, to enable a user to import an existing hierarchy  
from a given domain (animal taxonomy, ULAN thesaurus etc.)

I'd be happy to answer any question about SEM, if you have any, and/or  
to create mappings from SEM to EON, to ensure interoperability of  
datasets modeled with both models :)

Hope you'll find these pointers interesting!

Best,
Véronique

_______________________________________________
foaf-dev mailing list
foaf-dev-RyYwo1q5J+qsOXdr9/[email protected]
http://lists.foaf-project.org/mailman/listinfo/foaf-dev
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.