Date: Mon, 08 Jun 1998 16:20:41 -0500
From: Gerry Mckiernan <GMCKIERN@GWGATE.LIB.IASTATE.EDU>
Subject: Content Analysis of Academic Departmental Homepages

_Content Analysis of Academic Departmental Homepages_

As a science reference librarian and bibliographer [Yes, I do have a real
job, (and a wife, three kids, mortgage, etc. [:->], I seek to maintain an
understanding of the information needs of my clientele as well as the
research interests of the faculty and staff I serve in the selection of
materials that will best support their research interests.

Last year, in an effort to gain a better understanding of such interests,
I identified and reviewed the Web pages of my liaison departments as well
as the official individual homepages of each member of a department, e.g.
Aeronautical Engineering. My immersion and digestion of this formal
information, has been of great benefit in identifying and selecting
materials for purchase for our library collection [This knowledge has been
most useful in decisions relating to the purchase of more expensive
engineering monographs and proceedings].

At one point, I considered tabulating the interests of my departmental
faculty into a spread- sheet as a formal aid for assisting in the decision
process for retrospective purchases as well as future considerations.
However, in a recent revisit to my proj ect on the use of Intelligent
Software Agents for library applications, it occurred to me that an ideal
application of Agent Technology for collection development for libraries
would be one in which Agents analyze the contents of departmental
homepages and generate a group user profile department based upon a
synthesis of the expressed (and possibly implicit or latent) collective
research interests.

With such a collective user profile, one could now consider using it as a
Mega Search Statement that another agent would use to search the Web,
local (or remote) licensed index and abstract databases, other OPACs, etc.
to identify relevant resources for s ubsequent consideration for selection
and purchase and/or incorporation within the local 'collection'. Of
course, we would want the ability to instruct the Content Agent so that we
would be able to be selective in a choice of a department an/or to specify
 the type of electronic database for a subsequent search by the Search
Agent.

One would of course wish to manage that agents such that one could massage
the results of each agent such that results could be organized according
to professional judgment.

It would be hope that the results could in turn be used to identify the
deficiencies of the local 'collection'. For example, to identify those
e-journals that best 'suit' the interests of a department, or to identify
key Web resources that would serve the
 interests of a department or a research group within a department.

One could also imagine providing an alerting service to which a faculty
member could subscribe that would provide them with a Mega Current
Awareness Service of newly discovered items. [One could indeed consider
using another agent, a Feedback Agent, that in turn could provide a Real
Time update to each and every faculty members interests based upon their
selection and use of selected resources]

In planning for the formal establishment of my clearinghouse devoted to
the use of Agents for collection development, reference as well as
technical services, called _LibraryAgents(sm)_ {:->], I would be
interested in learning about any efforts envisioned , as well as those
related to it. BTW: The address for LibraryAgents(sm) is:

http://www.public.iastate.edu/~CYBERSTACKS/Agents.htm

Currently, this site has a fuller description of the Larger Project, as
well as links to key Agent clearinghouses.

[I am aware of the various e-mail alerting services offered by publishers
(e.g., Elsevier, IOP) and information services (e.,g EBSCO, ISI) and would
appreciate learning about any compendium of such Agent-based services as
basic background for Library Agen ts(sm).]

As Always, Any and All citations, sources, contributions, critiques,
questions, concerns, comments, or queries are Most Welcome!

Joy!

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

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

"The Best Way to Predict the Future is To Invent It" Attributed to Peter
Drucker