Ice-Cube(sm): Visual Representation and Navigation of LCSH Facets

[Jean P. Gordon]

What is facet analysis?

Jean Gordon

Gerry McKiernan wrote:
> 
>                        _Ice-Cube(sm):_
>      Visual Representation and Navigation of LCSH Facets
> 
>     In a recent posting [I believe it was this week],
> I inquired about any effort that has sought to perform
> a Facet Analysis on the Library of Congress Subject Headings
> (LCSH). Those who have responded to this query indicated that
> they were not aware of any current effort.
> 
>     With the expectation that this will be done some
> way, some day, it has occurred to me that there will be
> a need to represent and navigate the Facets once they
> are identified. For this awesome task, I propose
> the creation of a Information Space Cube [Let's call
> it Ice-Cube(sm) [:->]
> 
>     From a Facet Analysis, one could expect that every
> LC subject heading could be assigned to every and all
> facets to a matter of degree. A particular subject
> will certainly be a member of more than one facet category
>  [Of of course, a particular facet would also be a member of
> 
> more than one higher facet]. For the sake of argument
> let's assume that ten major facets were established for
> each LC subject heading [Below the Meta-Facet level, there
> of course would be SubFacet levels in which a specific
> subject heading would hold membership; again let's say
> that at the first SubFacet level there are ten sub-facet
> categories [What can I say, I'm feeling digital today
> [hey that rhymes [:->]]
> 
>    At the Meta-Facet level, a specific subject would
> have the possibility of having a value assigned to each
> facet for that level [Let's say, we discover through
> Algorithmic Magic, that these values can range from
> One (1) through 10 (ten) [Boy, Am I Digital] and that
> each assigned value represents the relative semantic
> weight that the particular subject has for a given facet
> at a given Facet
> 
> [If You Can Follow this Logic, Read-On, Otherwise Press 'Delete']
> 
>   Thus, for example, a specific subject heading, would
> be represented by a statement that consisted of a set
> of relative weights [Zero to 9] for ten facets [in facet order of
> course [:->]]
> 
> [If You Can Follow this Logic, Read-On, Otherwise Press 'Delete']
> 
>   The Ice-Cube(sm) would be a construct consisting of each
> of the ten facets on each the X axis, Y axis and the Z axis.
> A given subject heading would be located within Ice-Cube(sm)
> at the coordinates of the relative weight of the subject
> for each Facet at a given facet level [Weights could vary from
> Absolute Zero to 9.9999 etc. [We'd need to set a threshold -
> up to 10 decimal places [Oops, there I go again Being Digital
> [With apologies to Negroponte] [Hey, Decimals are Digital![:->]
> Other LC Subject Headings with a similar facet statement
> would have a similar location within the Ice-Cube(sm).
> Hopefully, [if we make the appropriate sacrifices to the
> Automated System Gods and Goddesses and the Comp Sci Gurus]
> subject headings with the same facet profile [and hopefully
> 'meaning' [whatever that may be in the Faceted Universe]
> would be clustered together for subsequent navigation.
> 
> [If You Can Follow this Logic, Read-On, Otherwise Press 'Delete']
> 
>   Of course, we need to have different Ice-Cube levels to
> allow users to browse from the MetaFacet level to SubFacet levels
> [We'd need a sub-cube for each of the Meta-Facet categories
> indicating the relatively associations of sub-facets at
> the sub-facet level] [The Meta-Facet level would of course
> be called the Ice-Cream(sm) [Cream raises to the top - get
> it]
> 
>  [If You Can Follow this Logic, Read-On, Otherwise Press 'Delete']
> 
>   Ice-Cube(sm) would then be the interface for the display of
> the LCSH as an interface to an LCSH OPAC. Users would browse
> subjects by interacting with them within the Ice-Cube(sm).
> The interaction within a Faceted OPAC (a FacPac ? [:->]
> will be the topic of another posting in the near future
> [Can you wait?]
> 
>   [I was going to call this metaphor Ice-Cubed - Ice to the
> Third Power, but I don't have a superscript on this SGI Indy {:-(]
> 
>   Certainly, other metaphors can be considered for mapping
> and displaying the faceted essence of a subject heading,
> including those in my clearinghouse devoted to Information
> Visualization, _The Big Picture_ available at URL
> 
>    http://www.public.iastate.edu/~CYBERSTACKS/BigPic.htm
> 
>   I can see myself moving from my Cubist phase to some
> Out-of-the Box [:->] thinking to consider a spheroid metaphor;
> in that case I'd call it ... _The Whole Ball of Wax_ -
> it's certainly more global or at least globular {:->]
> 
> [If You Can Follow this Logic, Read-On, Otherwise Press 'Delete']
> 
>   All of the above sounds very mathematical. Can anyone
> tell me what branch of math I working in here. [I'd
> like to post to the appropriate Math groups two/too].
> 
>   But seriously folks, I'd very much appreciate any and all
> information about efforts to display and/or visualize
> facet associations, LCSH or otherwise.
> 
>   [Afterthought: How about the facet analysis of other
> thesauri]
> 
>   <h1> Thanks </h1>
> 
> Gerry McKiernan
> Curator, CyberStacks(sm)
> Iowa State University
> Ames IA 50011
> 
> gerrymck at iastate.edu
> http://www.public.iastate.edu/~CYBERSTACKS/
> 
>             "What's It All About, Alfie?"
> 
> [If You Can Follow this Logic, Read-On, Otherwise Press 'Delete']

-- 
*************************************************************
Jean P. Gordon				  Phone: 415-257-0196
Archbishop Alemany Library		  Fax:   415-459-2309
Dominican College
50 Acacia Avenue	       E-mail: gordonjp at dominican.edu
San Rafael, CA 94901	   Web page: http://www.dominican.edu
*************************************************************