Christoph, All,

Just to note literals for feature values might not be sufficient,
especially for data mining algorithms. (We had already identified the
need of data type with a recent discussion with  Tobias and Pantelis ).

There is an extensive support for data types in RDF

<?xml version="1.0"?>

<rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"

         xmlns:ex="http://example.org/stuff/1.0/">

  <rdf:Description rdf:about="http://example.org/item01">

    <ex:size
rdf:datatype="*http://www.w3.org/2001/XMLSchema#int*">123</ex:size>

  </rdf:Description>

</rdf:RDF>


  And the hierarchy of built-in data types according to W3C
  Recommendation 28 October 2004 за XML Schema Part 2: Datatypes Second
  Edition at http://www.w3.org/TR/xmlschema-2



Diagram of built-in type hierarchy


I'll send another email about feature definitions. If we consider them
replaced by predicates, these predicates needs still to be defined with
RDF , including, names, units, authors, references, etc.


Best regards,
Nina


Christoph Helma wrote:
> Dear all,
>
> If we agree on RDF for data exchange, we can write feature values as
> literals. This would make the feature API obsolete, because
>
> 	- the feature URI (predicate) substitutes the previous feature-definition
> 	- the feature value (object) substitutes the previous feature
>
> The remaining functionality of the original API is covered by the
> dataset component + RDF. Less is more!
>
> Best regards,
> Christoph
> _______________________________________________
> Development mailing list
> Development at opentox.org
> http://www.opentox.org/mailman/listinfo/development
>