Hi Egon,

Egon Willighagen wrote:
> The ontology just describes the entities involves... this is
> complemented by a dictionary of QSAR descriptors that uses the
> ontology to express the knowledge...
>
> Nina, I'd love to set up a dictionary of machine learning
> algorithms... let's continue on this.
>
>   
You are most welcome :) This is actually an urgent task we need to
complete at the beginning of January latest.
>> An initial version of algorithm types ontology is available here
>> http://opentox.org/data/documents/development/RDF%20files/AlgorithmTypes/view
>> .  Could you please have a look and then we will discuss where does your
>> algorithm fit?
>>     
>
> I'll have a look at that.
>
>   
>>> 4) There is no prior experience with RDF/OWL, we will need some help here.
>>>
>>>       
>
> I have quite some experience with RDF, and certainly assist. We are
> using RDF in Bioclipse, for example, for dynamically look up QSAR
> descriptor information, but you can check my chem-bla-ics blog for
> many more applications. I can recommend the following web page for
> questions about semantic web, OWL, RDF, etc:
>
> http://www.semanticoverflow.com/
>   
>   
There have been several concerns of the performance of services,
generating/parsing large amount of  RDF/OWL data.  It might be only
related to current implementation - for example Jena needs to build the
entire model in memory, prior writing it.  For comparison, Ambit
services are streaming SDF and ARFF files, but not RDFs currently, so
there indeed might be a performance penalty for large dataset.

Do you have experience with a streaming RDF parser - and possibly a
recommendation?

Best regards,
Nina
> Egon
>
>
>