Copy Link
Add to Bookmark
Report
IRList Digest Volume 3 Number 36
IRList Digest Sunday, 4 October 1987 Volume 3 : Issue 36
Today's Topics:
Query - Personal filing systems and work in cognitive psychology
- Social and technical means of storing knowledge in organizations
- Application of IR techniques to info. problems of undergraduates
- Availability of CODER lexicon based on Collins dictionary
Reply - Interest in parallel computation and IR
Announcement - Tenure track position at Syracuse University
COGSCI - Systems with Multiple Expertise
- Protein Structure and Pattern Matching
- Multilingual Semantic Structures from Machine-Readable Dictionaries
News addresses are
Internet or CSNET: fox@vtcs1.cs.vt.edu
BITNET: fox@vtcs1.bitnet
----------------------------------------------------------------------
Date: Mon, 28 Sep 87 19:06 PDT
To: Edward A. Fox <FOX@VTCS1>
From: Donald Case <IIN8PXB@UCLAMVS>
Subject: personal filing systems and work in cognitive psychology
. . .
I am starting a project that will examine possible connections between
personal filing systems and work in cognitive psychology on categorization.
To begin with, I will be studying historians, as they are knowledge workers
who must deal with massive amounts of textual material, in a field that has
some accepted structures (but is otherwise as broad as any field could be)
and who are not yet as automated as scholars in some other fields. I would
apprciate hearing from anybody who has similar interests or who knows of
kindred research. I believe that such studies could help us to devise more
natural and adaptive filing systems for personal workstations, as well as
helping us to better understand the nature of scholarhsip itself.
-- Donald Case, UCLA IIN8PXB@UCLAMVS
------------------------------
Date: Tue, 29 Sep 87 01:06 EST
From: STEIN@wharton.upenn.edu
Subject: social and technical means of storing knowledge in organizations
To: members of the IRlist.
From: Eric W. Stein
Systems Research Fellow
Department of Social Systems Sciences
The Wharton School
University of Pennsylvania
3733 Spruce Street
Philadelphia, PA 19104
(215) 898-4472
email: stein@wharton.upenn.edu or stein@wharton-10.arpa
Re: Request for citations and information regarding research being
conducted on the social and technical means of storing knowledge
in organizations.
The department to which I belong conducts research on social systems;
that is, on systems composed of people and technologies that are transformed
over time. The graduate group in social systems science owes its roots to the
the works of Russell Ackoff, Hasan Ozbekhan, Eric Trist, Fred Emery,
and Tom Saaty who were among the principle founders. Theoretical roots
lie in systems, cybernetics, planning, organizational design, decision
science, socio-technical design, and management theory.
I am developing a dissertation on the social and technical
means by which organizations (most likely in the private sector)
store core organizational knowledge including expertise,
values, procedures, and strategic knowledge about the environment.
I am looking at the technical means by which organizations accomplish
knowledge storage such as the implementation of expert systems as well
as the more subtle, social means such as through systems of
communication, myth-making, and participative planning to name a few.
I am interested in those who are working in this area; my review of the
literature has failed to produce many studies. Please
contact me if you know of important citations or models for dealing
with the general area mentioned; esp. the problems of mapping out
knowledge structures in organizational settings, the effects of
expert systems technologies on social systems, network
theories of organizations, and evaluation of
expert systems from the users point of view.
I hope I have provided you with enough information to capture
the attention of those who share the same interests. If even
mildly interested, feel free to contact me at the address listed.
Yours,
Eric Stein
stein@wharton.upenn.edu
------------------------------
Date: Mon, 28 Sep 87 12:32:47 CST
From: JEFF HUESTIS <C81350JH@WUVMD>
Subject: ELSI/VTLS/SMART, ETC.
To: "EDWARD A. FOX" <FOX@VTCS1>
I would appreciate receiving a brief update on the status of the
ELSI-SMART-VTLS connection, as mentioned in your March, 1986 article
in Microcomputers for Information Management. I'm particularly
interested in the application of IR techniques to the information
problems of undergraduates, for a number of personal, professional,
and political reasons. More specifically, I'm looking at ways to
take an unfocused browsing search, or an erroneously precise search
(eg., using the wrong terminology), and guiding it onto the beaten
path of what has been published and what is available in the local
library collection (given the limited interlibrary support usually
available to undergraduates).
[Note: Tom McGuire finished ELSI in connection with his Masters work
here. It ran on a 3B2 which AT&T had been kind enough to donate.
Since then we were able, thanks to the kindness of the Virginia Tech
library staff and personnel at VTLS, Inc., to obtain virtually all of
the catalog data as of 12/85. We ran a preliminary experiment with a
version of the SMART system we extended to handle the data, using some
300,000 catalog entries -- an article discussing some of the findings
should appear soon in "Information Services and Use." We hope to
extend that study, using some 600,000 catalog entries, later this AY. - Ed]
One approach I'm taking is a kind of crude co-citation-based clustering
of 14-years worth of circulation archival data from one of the central
library's old batch systems, consisting of about 1.5 million charge
records. Each record includes the LC call number, very brief author
title information, and the borrower number, among other fields.
Working from the assumption that most borrowers check out books on
loosely related topics within a delimited period of time, I've clustered
these transactions into borrower/time-period records which are indexed
by the call numbers of the transactions in each record. At present,
the time-period represented by each record is one academic year, though
I'm planning to reorganize the file into quarterly records, at least
for the heaviest borrowers.
The primary value of this file, beyond the
title/user/title linkage pathways established (analogous to the
combined use of bibliographies and citation indexes), might be to
extend the browsing support provided by the LC classification, by
providing possible cross-references between arbitrarily small subtrees.
But this really depends on cleaning up what is now a very dirty file,
eg., by the use of statistical techniques to separate the records of
focused researchers from those of renaissance men.
...
[Note: This reminds me of the study by Paul Metz of Virginia Tech, in
the ALA book that is ACRL Publications in Librarianship no. 43, "The
Landscape of Literatures: Use of Subject Collections in a University
Library." - Ed]
[Note: Following are more comments from Jeff after he read my note:
"I read Metz's book cover-to-cover, which is unusual for me. Its
primary orientation was toward collection development, a sort of
macro-view. But I found a lot that was relevant to the problem of
retrieval, the micro-view (borrowing from the macro/micro distinction
from economics)."]
--Jeff Huestis
c81350jh@wuvmd.bitnet
Washington University Libraries, St. Louis
[Note: I encourage other comments and discussion on this topic - many
of the IRList readers are involved in library science or library work
of one type or another and do not often send in news or comments - Ed]
------------------------------
Date: Wed, 12 Aug 87 15:55 N
From: <COR_HVH@HNYKUN52>
Subject: coder dictionary
Is the CODER lexicon available to the outside world, i.e. can one get a
copy of it (and how)?
I am interested for two reasons in your results.
1) The PROLOG version of Collins could be of use to our analysis efforts.
2) If you have by now actually parsed the adverb definitions, you have
a bunch of tree structures I could put in my database (the Linguistic
Database I am always talking about) as yet another demo application.
Hans van Halteren (Nijmegen University, Netherlands)
[Note: I am assured by my graduate students and secretary that a tape
will go out to Oxford Text Archive tomorrow with the CODER version of
the Collins Dictionary of the English Language. The contents of the
tape could be directly loaded into a (large) Prolog fact base, or with
some minor changes, could probably be put into a relational database.
It should be possible for you and others to obtain a copy of the tape
by going through the normal channels of requesting the dictionary from
OTA, for use in research.
Regarding the parsing question, we are not parsing definitions
here now. The NSF has funded a joint project at Illinois Institute of
Technology and Virginia Tech, part of which will deal with parsing
Collins definitions, but we are won't have those results for quite
some time. - Ed]
------------------------------
Date: Thu, 24 Sep 87 13:45:24 edt
From: ltw (Layne Watson)
Subject: Re: IRList Digest V3 #35
To: Keith van Rijsbergen cjvr@cs.glasgow.ac.uk
Several of us here at Virginia Polytechnic Institute & State University
have interests in both parallel computation and IR. My background is
primarily in scientific and parallel computing, but I have recently
become interested in IR as it relates to supercomputing and parallel
computing. We are trying to acquire parallel hardware, and have lots of
ideas we want to try out.
Layne Watson
Department of Computer Science
Virginia Polytechnic Institute & State University
Blacksburg, VA 24061 USA
watson@vtcs1.bitnet or watson@vtcs1.cs.vt.edu
------------------------------
Date: Mon, 28 Sep 1987 12:08:39 LCL
From: RNODDY@SUVM
Subject: Job Opening at Syracuse University
. . .
SYRACUSE UNIVERSITY - SCHOOL OF INFORMATION STUDIES
ASSISTANT/ASSOCIATE PROFESSOR - TENURE TRACK
The School is seeking an individual for a new full-time, tenure-track
faculty position, beginning September 1988. The primary responsibilities
for this position are to teach and conduct research in the areas of:
* Information Retrieval
* Natural Language Processing
* Knowledge Representation
* Artificial Intelligence
* Application of Parallel Architectures to Information Science
Applicants should be committed to excellence in teaching and research, and
be interested in working with an interdisciplinary faculty to apply
their expertise to the current theoretical and applied problems of our
field. PhD is required, though candidates near completion are encouraged
to apply.
In December 1988, the School will be housed in a new Science and
Technology Center for the express purpose of engaging in interdisciplinary
research with the Schools of Engineering and Computer Science, as well as
the Northeast Parallel Architecture Center (NPAC), and the Computer
Applications and Software Engineering Center (CASE).
The closing date for applications is December 15, 1987. Applications
(including letter, resume and names of three references) should be
addressed to:
Dr Jeffrey Katzer, Chair, Search Committee
School of Information Studies, Syracuse University
200-K Huntington Hall
Syracuse, New York 13244-2340
Syracuse University wholeheartedly supports the Affirmative Action Program.
(Preliminary contacts by potential applicants will be welcomed by
Jeffrey Katzer or Robert Oddy at (315)423-4525, or
Robert Oddy at rnoddy@suvm on BITNET.)
------------------------------
Date: Mon, 28 Sep 1987 10:34 EDT
From: Peter de Jong <DEJONG%OZ.AI.MIT.EDU@XX.LCS.MIT.EDU>
Subject: Cognitive Science Calendar [Extract - Ed]
Date: Thursday, 24 September 1987 17:06-EDT
From: Marc Vilain <MVILAIN at G.BBN.COM>
Re: BBN AI Seminar -- Alexander Makarovitsch
BBN Science Development Program
AI/Education Seminar Series
SYSTEMS WITH MULTIPLE EXPERTISE
Alexander Makarovitsch
Computer Science Department, West Catholic University
(Angiers, France)
BBN Laboratories Inc.
10 Moulton Street
Large Conference Room, 2nd Floor
10:30 a.m., Tuesday, September 29, 1987
Abstract: By a system with multiple expertise, we mean one containing
two or more expert systems, all working on a common information domain.
Three such systems under current development will be briefly reviewed.
(1) Syclope, a system being developed for the French Board of
Electricity for aiding agents working under risky conditions to
improve their task behavior;
(2) Sonar, a system being developed for the Bull Computer Group for
aiding in the design, operation, and maintenance of computer
networks;
(3) Link, a system being developed for the Bull Group for aiding
decision makers in the networked computer area in the processes of
product planning.
The review will focus on the difficulties encountered in key aspects of
the development process: the acquisition of expertise, knowledge
representation, man/machine interface, and interactions amopng the
system and multiple users and experts.
------------------------------
Date: Wed, 30 Sep 1987 12:05 EDT
From: Peter de Jong <DEJONG%OZ.AI.MIT.EDU@XX.LCS.MIT.EDU>
Subject: Cognitive Science Calendar [Extract - Ed]
Date: Tuesday, 29 September 1987 12:08-EDT
From: Fran Lewitter <lewitter at PROPHET.BBN.COM>
Re: Biotech Sem - 10/1/87
Biotechnology Seminar Series
BBN Laboratories
Science Development Program
Topic: Protein Structure and Pattern Matching
Speaker: Dr. Robert M. Abarbanel
Technical Manager KEEconnection and IntelliScope
Intellicorp
Location: Room 3/130 (Weiner Room)
50 Moulton Street
Time & Date: Thursday October 1, 1987, 11:00 a.m.
Abstract: Pattern matching has been applied to the problem of assigning
protein secondary structure with some success. This talk will
discuss the ideas implemented using a pattern matching language
called PLANS, for finding secondary structures and turns in
globular proteins. These assignments are then extended to
determine reasonable tertiary models for several proteins. In
particular, work with Interleukin-2 and Human Growth Hormone will
be discussed. The Macro-Molecular Workbench project at UCSF will
also be mentioned.
Note: For more information about this seminar series, please contact
Fran Lewitter, Life Sciences/Research Systems - x2751. Please
note that following the seminar we will meet for lunch and
further discussion in the 7th floor cafeteria, 10 Moulton Street.
------------------------------
Date: Fri, 2 Oct 1987 12:19 EDT
From: Peter de Jong <DEJONG%OZ.AI.MIT.EDU@XX.LCS.MIT.EDU>
Subject: Cognitive Science Calendar [Extract - Ed]
Date: Thursday, 1 October 1987 16:40-EDT
From: mvilain at G.BBN.COM
Re: BBN AI Seminar --John White
A METHODOLOGY FOR
DERIVING MULTILINGUAL SEMANTIC STRUCTURES
FROM MACHINE-READABLE DICTIONARIES
John S. White
Martin Marietta Corp.
BBN Laboratories Inc.
10 Moulton Street
Large Conference Room, 2nd Floor
4:00 p.m., Thursday, October 8, 1987
This paper describes a methodology for generating multilingual
taxonomic hierarchies for specific and general terminological systems,
using the lexical relations derivable from machine-readable
dictionaries (MRDs). Taxonomic structures can be optimized to form
the data foundation of a knowledge base facilitating the automatic
disambiguation of text in one language into lexical senses of terms in
the same or another language. Such a capability, in turn, contributes
a multi-directional translation dimension to automatic paraphrase and
content analysis. As part of a full natural language processing
machine translation system, the knowledge added by the MRD structures
can assist in sense selection, grammar rule-base composition, and
generation.
The methodology for the development of these structures is described
here, consisting of a manual pre-disambiguation task, and
computational creation of parallel taxonomic structures in each
language. The rationale behind deriving lexical semantic structures
in this way is discussed.
------------------------------
END OF IRList Digest
********************
