Re: Best Practices - Semantic Tagging

HI Paolo,

On Sat, Mar 2, 2013 at 3:12 PM, Paolo Ciccarese
<paolo.ciccarese@gmail.com> wrote:
> Dear Tim,
> Apologize, I sent the email by mistake and I was still in the process of
> editing.
>
> I just wanted to add that I don't recall any discussion close to what you
> are asking.
> So what I expressed was purely my personal first reaction to your problem.
>

Thanks.  I am actually looking for personal reactions.  I don't recall
seeing anything "exactly" like this before either.
For conciseness; I am really leaning towards:

<xs:complexType name="ct-fd897186-1ec9-4534-bdca-3be434c93924"
rdfs:isDefinedBy="http://purl.bioontology.org/ontology/SNOMEDCT/365761000">

My reasoning is:
1) there are no pre-existing approaches, that I have found
2) it is very concise
3) it ties the complexType name directly to the semantic attribute
4) it is legal in XML Schema
5) it is easy to query the name and semantic link type together,
irregardless of which attribute the knowledge modeller decides to use.

Not shown before but the element in the instance and the complexType
name share the UUID. So we have:
<xs:element name="el-fd897186-1ec9-4534-bdca-3be434c93924"
type="mlhim2:ct-fd897186-1ec9-4534-bdca-3be434c93924"/>

Since they are all unique, it is easy to take the element name and
look it up in the collection of schemas.

Thoughts?  Again, previous uses cases and/or personal opinions are
encouraged.  :-)

> Also the use of oa:SemanticTag to fit into XMLSchema has to be thought
> through as I think that is far from what the term has been conceived for.

This is helpful information.  Thank you.

Cheers,
Tim


============================================
Timothy Cook, MSc           +55 21 94711995
MLHIM http://www.mlhim.org
Like Us on FB: https://www.facebook.com/mlhim2
Circle us on G+: http://goo.gl/44EV5
Google Scholar: http://goo.gl/MMZ1o
LinkedIn Profile:http://www.linkedin.com/in/timothywaynecook

Received on Sunday, 3 March 2013 00:09:36 UTC