Neo-Conventional' Thesauri

[Gerry McKiernan]

               _'Neo-Conventional' Thesauri_

    In my recent posting on structured browsing for
'user-controlled' information retrieval [Is Precision
Too Precise?] and a follow-up on the application of
Natural Language Processing (NLP) for Library of Congress
Subject Headings (LCSH), I make note of the (potential)
usefulness of such 'neo-conventional' approaches for
facilitating browsing of Information Spaces, most notably
with a database of MARC records (e.g. an OPAC).

    In considering the extensive of thesauri for bibliographic
databases, it also seems relevant to consider the potential
usefulness of 'navigating subjects' by applying NLP to
controlled vocabularies. The hope here is to reveal [i.e. to
permit users to discover] 'neo-conventional' relationships
among terms and phrases within this structured vocabulary
that is not offered by the syndetic structure of the cross
references. Certainly Data Mining and Knowledge Discovery in
Database technologies could also identify possible relationships
with these controlled vocabularies as well. Latent Semantic
Indexing (LSI) might also permit one to discover other kinds
of associations not unearthed by NLP or DM or KDD.

    For my never-ending review of Data Mining and Knowledge
Discovery in Databases (KDD), I would very appreciate learning
of efforts or systems that have applied such methods as NLP
LSI, DM/KDD, etc. to thesausri. I am aware of the highly-innovative
work of Harter (Indiana), Jones et al. (City University, UK),
Johnson (Illinois) and of course, doszkocs at NLM.

   As always, any leads, suggestions, citations, opinions,
comments, critiques, criticisms, door prizes [:->], etc.
are most welcome!

   Regards,

Gerry McKiernan
Curator, CyberStacks(sm)
Iowa State University
Ames IA 50011

gerrymck at iastate.edu
http://www.public.iastate.edu/~CYBERSTACKS/

                  "Show Me The System!"