Re: SKOS and MeSH qualifiers

In message 
<F5839D944C66C049BDB45F4C1E3DF89DEE9E59@exchange31.fed.cclrc.ac.uk> on 
Mon, 11 Jul 2005, "Miles, AJ (Alistair)" <A.J.Miles@rl.ac.uk> wrote
>I think I understand the problem (please correct me if I haven't), how 
>about this for a solution ...
>
>The fundamental statement that is being made in document metadata is 
>something like, 'the subject of document X is (concept Y qualified by 
>concept Z)'.
>
>The crucial bit is the parentheses.

Al -

This is the statement that is being made when you apply a single MeSH 
heading with a qualifier. The solution you propose may therefore work OK 
in this specific case. Other MeSH headings, with or without qualifiers 
can be applied separately to the same document to reflect other subjects 
that it deals with.

If that is done, the indexing is hybrid, with some pre-coordination 
(term + qualifier) and some post-coordination (other headings applied 
separately). That is the normal way in which MeSH is used.

I was more concerned to see whether SKOS could provide a generic 
solution to accommodate the more general case of pre-coordinated strings 
of indexing terms, of which the MeSH case would be a specific, more 
restricted, application.

In the more general case, a pre-coordinate string contains several terms 
(i.e. concepts) which together express the compound subject of a 
document. There may be various types of relationship between the 
concepts that are brought together in such a string; these relationships 
are not normally expressed explicitly, though they may affect the order 
in which the terms are cited within the string. In general none of the 
terms need be considered as "qualifying" any other - they are brought 
together as equals.

Examples of compound strings (subject headings taken from the Library of 
Congress catalogue) are:

Leukemia -- Animal models -- Congresses.
Leukemia -- Biography -- Fiction.
Leukemia -- Chemotherapy -- Data processing.
Leukemia -- Environmental aspects -- Massachusetts. Cape Cod.
Leukemia -- Patients -- Australia -- Biography.

With other schemes of  subject headings and classification (such as UDC, 
Bliss, and to a lesser extent Dewey) you may need to express 
pre-coordinated subjects with even more component concepts, such as:

"manufacture of packaging products from recycled waste paper"
"local authority provision of social services for families of 
disadvantaged children"
"user studies for the investigation of the effectiveness of computerised 
information retrieval systems"

I am not saying that SKOS _should_ try to accommodate complex 
pre-coordinated strings, though I believe that I did see somewhere an 
aspiration to cover classification systems as well as thesauri. Perhaps 
someone could clarify how far down this road you think it appropriate to 
go. I am just sounding a warning that if you adopt an ad-hoc solution to 
deal with the restricted case of "descriptor - qualifier" needed for 
MeSH, this should not be done is such a way as to be incompatible with 
any more general solution that you may wish to work towards in future.

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 Monday, 11 July 2005 22:32:06 UTC