W3C home > Mailing lists > Public > public-esw-thes@w3.org > May 2006

Re: AAT and SKOS

From: Leonard Will <L.Will@willpowerinfo.co.uk>
Date: Fri, 12 May 2006 21:58:47 +0100
Message-ID: <yATCuMIHcPZEFAkZ@willpowerinfo.co.uk>
To: public-esw-thes@w3.org

In message 
<DEF32271883F8941B74CE60E7388C5F60377BB@diocletian.amplexor.com> on Fri, 
12 May 2006, Paul Hermans <paul.hermans@amplexor.com> wrote
>Dear all,
>I'm trying to map the AAT thesaurus to SKOS.
>In AAT I do find 4 types of records:
>C – Concept: This is the VCS default value when you create a new 
>record. Most terms in the AAT are called "concepts," including all 
>terms that can be used for indexing, that is all terms that are not 
>guide terms, "hierarchy" names, or facets. If the record you are 
>creating is NOT concept, change the Record Type to an appropriate value.
>H – Hierarchy name: Refers to the top of a hierarchy. The hierarchy 
>name is not used for indexing or cataloging.
>G - Guide Term: A Guide Term (also called a node label) is a level used 
>to organize the hierarchy into logical segments. In AAT, Guide Terms 
>are used only with historical hierarchies and candidate hierarchies. 
>Consult with your supervisor before using this designation.
>F – Facet: A facet is a broad division of the hierarchy, generally 
>appearing directly under the Root
>C - Concept maps cleanly to SKOS concept
>G - Guide term, maps to SKOS collection
>F - Facet can be defined as a subclass of concept.
>But I'm not sure what to do with the record of type "Hierarchy name".
>Do I map this also onto a collection, but how to distinguish it then 
>from guide term collections?

In my opinion, the AAT confuses some of this labelling, and is not 
entirely consistent.

I agree that facet can be defined as a subclass of concept. The names of 
facets, such as "objects", "activities", "materials" and so on are 
indeed concepts, though so broad as unlikely to be useful as index 

As far as I can see, the "hierarchy name" is just another concept which 
happens to be the first-level subdivision of a facet, though the AAT 
calls it "the top term of a hierarchy", and I don't see why it should 
not be used for indexing if it is the most appropriate term to express 
the subject of a document (or "resource" if you prefer). Within the 
"objects" facet there are hierarchy names such as "Built Environment", 
"Furnishings and Equipment", and "Visual and Verbal Communication 
[artifacts]", and these seem perfectly reasonable terms to express these 
broad concepts. They might be needed to index documents which are broad 
treatments of these concepts or to index documents on specific aspects 
which have not been provided for in the enumerated terms.

Similarly, the AAT combines two quite different kinds of thing into what 
it calls "guide terms":

a. Some of these are just terms for concepts that are needed to provide 
a useful step in the hierarchical structure, such as <recreational 
artifacts>, <sports and athletic equipment> or <air and space 
transportation vehicles>. I think that these should be treated as 
indexable concepts, perhaps with a usage note "Use a more specific term 
if possible" (though as that is a guideline that should apply everywhere 
it may be thought redundant).

b. Other things that the AAT calls "guide terms" are indeed node labels, 
specifying a characteristic of division by which the following array has 
been grouped. These contain the word "by", as in <toys by form>, <toys 
by function>, <sound devices by acoustical characteristics> and so on.
(I see that SKOS calls these "collections", though I find that term to 
be rather confusing, because we are not dealing with collections or 
groups of concepts (or of documents or resources, as some people might 
be mislead into thinking) but of ways in which members of a broader 
concept can be differentiated and split into narrower concepts.)

The issue in converting the AAT to SKOS is then whether you follow the 
AAT's designation of the different elements in its structure or whether 
you treat elements in the way I have indicated above, which I believe is 
more logical, irrespective of what the AAT calls them.

Leonard Will

Willpower Information       (Partners: Dr Leonard D Will, Sheena E Will)
Information Management Consultants              Tel: +44 (0)20 8372 0092
27 Calshot Way, Enfield, Middlesex EN2 7BQ, UK. Fax: +44 (0)870 051 7276
L.Will@Willpowerinfo.co.uk               Sheena.Will@Willpowerinfo.co.uk
---------------- <URL:http://www.willpowerinfo.co.uk/> -----------------
Received on Friday, 12 May 2006 20:59:27 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 2 March 2016 13:32:07 UTC