NIF produces blank node free RDF, which is simple to understand and
flexible to use:
http://svn.aksw.org/papers/2012/WWW_NIF/public/string_ontology.pdf
The basic idea is to model Strings in Context.
So you have a document:
http://www.w3.org/DesignIssues/LinkedData.html
And you want to refer to its content (all characters from 0 to the end ):
http://www.w3.org/DesignIssues/LinkedData.html#offset_0_26546
Now you define the content as the Context and say, that it occurs in
the Document:
@prefix : <http://www.w3.org/DesignIssues/LinkedData.html#>
:offset_0_26546 a str:Context ;
str:occursIn <http://www.w3.org/DesignIssues/LinkedData.html> .
Additionally you can say that a substring occursIn a paragraph:
:offset_600_1000 a str:String ;
str:referenceContext :offset_ 0_26546 .
str:occursIn <yourparagraphuri> .
Or you can model each paragraph as a different context:
:offset_600_1000 a str:Context ;
str:occursIn <yourparagraphuri1> .
:offset_1001_2000 a str:Context ;
str:occursIn <yourparagraphuri2> .
The only drawback is that there is only an old owl ontology for NIF
1.0 http://nlp2rdf.org/nif-1-0 . NIF 2.0 will not be backwards
compatible:
nlp2rdf.org/news/nif-roadmap-2012-and-pointers
We hope to have it ready in two months.
All the best,
Sebastian
Zitat von Leyla Jael García Castro <[email protected]>:
Hi all,
I am modelling the structure and content of document. I am using DoCO (
http://purl.org/spar/doco<http://www.essepuntato.it/lode/http://purl.org/spar/doco>)
for the structure, i.e. Section, Paragraph, etc.
Now I need to model the actual content in a paragraph, meaning the text in
it. Here an example:
<myDocumentURI> pattern:contains [ a doco:Section
; pattern:contains [ a doco:Paragraph
; -content property- "This is the text in the paragraph." ] ] .
What would be the best approach to do that? *bibo:content* (
http://bibliontology.com/) is deprecated so I do not want to use that one;
the recommendation there is to use *rdf:value*, that would be an
alternative. *sioc:conten*t is another alternative. It could also be *
c4o:hasContent* (http://purl.org/spar/c4o). I guess there are also other
alternatives out there.
Which one would be the best alternative? Any suggestions?
Thanks so much,
lj
----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.
