RE: IVOA Thesaurus

From: Alasdair Gray <agray-at-dcs.gla.ac.uk>
Date: Tue, 30 Oct 2007 16:28:25 -0000


Hi,

>> First of all I would like to talk about top level concepts. I found  
>> that there were 2,646 top level concepts. I feel this very high,  
>> particularly as there are only about 2,850 concepts in total. This  
>> is also particularly bothersome if you analyse the IAU Thesaurus  
>> which only has 516 top level concepts.

> Strange, my python script claims that there are 2972 tokens in the
> newest version (which now includes all the elements and so had to
> grow a little bit!) and only 968 top concepts. The script simply
> looks for BT's and NT's, which we know are not yet internally
> consistent, so maybe this is a good time to clean up the ontological
> connections. Alasdair undoubtably used some smarter tools than my
> python script and the RDF hasTopConcept list should be taken with a
> grain of salt...

I'm afraid to say that I did not use anything particularly clever, just a few greps with wc -l. Also I counted the hasTopConcept to find the number of top level concepts. Rick, I've not looked closely at your script, but why is it generating so many top level concepts while counting a smaller number?

> A more formal question: I've been shoving UF's into aliases (a
> process which has not been completed, since I'm mostly doing this
> myself on spare time), but SKOS doesn't seem to have a means of
> expressing U's?

The U and UF should be modelled as skos:altLabel and skos:prefLabel respectively. I'll illustrate this with an example from the original IAU thesaurus:

acceleration of gravity
U

	GRAVITY
	SURFACE GRAVITY

Should result in both the gravity concept and surface gravity concept having the declaration

        <skos:altLabel>Acceleration of gravity</skos:altLabel>

Specifically, there should not be a concept for acceleration of gravity.

I hope this clears the issue up for you Rick,

Alasdair

-----Original Message-----
From: owner-semantics-at-eso.org [mailto:owner-semantics-at-eso.org] On Behalf Of Frederic V. Hessman
Sent: 30 October 2007 09:55
To: IVOA semantics
Subject: Re: IVOA Thesaurus

Alasdair Gray noted:

> First of all I would like to talk about top level concepts. I found
> that there were 2,646 top level concepts. I feel this very high,
> particularly as there are only about 2,850 concepts in total. This
> is also particularly bothersome if you analyse the IAU Thesaurus
> which only has 516 top level concepts.
Strange, my python script claims that there are 2972 tokens in the newest version (which now includes all the elements and so had to grow a little bit!) and only 968 top concepts. The script simply looks for BT's and NT's, which we know are not yet internally consistent, so maybe this is a good time to clean up the ontological connections. Alasdair undoubtably used some smarter tools than my python script and the RDF hasTopConcept list should be taken with a grain of salt...

Latest changes:

The new documentation is a bit easier to use - just one top-level page at         

http://www.astro.physik.uni-goettingen.de/~hessman/rdf/IVOAT/

with links to the introduction, dictionary, and lists of tokens, labels, and aliases. The alphabetic link buttons at the bottom aren't perfect but should enable you to browse more conveniently (the direct links between the list items and the dictionary are now broken - next thing to be fixed).

A more formal question: I've been shoving UF's into aliases (a process which has not been completed, since I'm mostly doing this myself on spare time), but SKOS doesn't seem to have a means of expressing U's?

Rick

--

Alasdair J G Gray
Research Associate: Explicator Project
http://explicator.dcs.gla.ac.uk
Computer Science, University of Glasgow
0141 330 6292
Received on 2007-10-30Z17:28:41