Re: Linked Data for ICD? OPCS-4?

On Sun, Oct 4, 2009 at 4:10 PM, Hugh Glaser <hg@ecs.soton.ac.uk> wrote:

> On 04/10/2009 23:48, "Oktie Hassanzadeh" <oktie@cs.toronto.edu> wrote:
>
> >
> > On Sun, Oct 4, 2009 at 3:28 PM, Hugh Glaser <hg@ecs.soton.ac.uk> wrote:
> >> On 04/10/2009 23:18, "Peter Ansell" <ansell.peter@gmail.com> wrote:
> >>
> >>> 2009/10/5 Hugh Glaser <hg@ecs.soton.ac.uk>:
> >>>> Sorry, should have given an example:
> >>>> http://www4.wiwiss.fu-berlin.de/drugbank/page/drugs/DB00588
> >>>
> >>> There is also the issue that the dbpedia URI's are not the same as the
> >>> ones everyone else is using. Ie, http://www.dbpedia.org/ URI's are
> >>> quite distinct from http://dbpedia.org/ URI's that everyone else
> >>> including DBpedia use. A similar URI structure situation exists with
> >>> some erroneous references to symbol.bio2rdf.org <
> http://symbol.bio2rdf.org>
> >>> that were a temporary
> >>> experiment that has now been fixed to reference bio2rdf.org
> >>> <http://bio2rdf.org>  instead.
> >>> http://www4.wiwiss.fu-berlin.de/diseasome/page/genes/AAS
> >>>
> >>> These mappings are all defined in a configuration file right?
> >>> Shouldn't be too hard to fix these and reload the server?
> >> Possibly.
> >> But it doesn't make much difference, since linkedct has the same things:
> >> http://data.linkedct.org/page/intervention/5580
> >> owl:sameAs
> >> http://data.linkedct.org/page/intervention/5633
> >> Etc.
> >> because they are both owl:sameAs
> >> http://www4.wiwiss.fu-berlin.de/drugbank/resource/drugs/DB00588
> >>
> >
> > Both the interventions above talk about the same drug,  Fluticasone
> Propionate
> > (http://data.linkedct.org/page/intervention/5633) and Flixotide
> > (http://data.linkedct.org/page/intervention/5580) are the same drug
> according
> > to our source of semantic knowledge about drug names.
> >
> > But, the interventions are different entities (they might have different
> dose
> > or other attributes) that's why they are not linked with owl:SameAs. On
> the
> > LinkedCT side, they are linked to the related drugs on drugbank with
> > rdfs:seeAlso.
> Hi Oktie,
> As they should be.
> My abject apologies - I could have sworn they had sameAs have an hour ago.
> My eyes are getting old.
> You haven't just changed them have you? :-)
>


Actually I did change one of the mappings, just one that I had missed before
for some reason, right after your first message. But I swear it wasn't after
you sent the URIs :) You probably have checked the URIs before my update.
I'm sorry I should have mentioned this in my message, but anyway I didn't
mean to say that what you said was wrong, I just wanted to clarify the issue
regarding LinkedCT interventions, and that there are still sameAs links on
the other side that should be fixed. Your point is still valid until the
other side is fixed since the sameAs link can be inferred indirectly anyway.

Thanks,
Oktie



> Best
> Hugh
> >
> > On the drugbank side however these links should be fixed, also the links
> to
> > dbpedia and bio2rdf should be fixed as pointed out by Peter. I'm CC'ing
> Anja
> > who maintains the dataset, hopefully she can fix it soon.
> >
> >
> > Regards,
> > Oktie
> >
> >
> >
> >
> >
> >> I love the Semantic Web :-)
> >>
> >> Cheers
> >> Hugh
> >>>
> >>> Cheers,
> >>>
> >>> Peter
> >>
> >>
> >
> >
>
>