nl-kr-request@cs.rpi.edu (NL-KR Moderator Chris Welty) (02/22/89)
NL-KR Digest (2/21/89 21:04:12) Volume 6 Number 4
Today's Topics:
CSLI Calendar, February 2
The PRATTFALL Machine-Translation Module (Unisys AI Seminar)
Query on NL-related MAIL-ADDRESSES
Parsing Greek
Submissions: nl-kr@cs.rpi.edu OR nl-kr@turing.cs.rpi.edu
Requests, policy: nl-kr-request@cs.rpi.edu OR
nl-kr-request@turing.cs.rpi.edu
%% The announcements here are old.
----------------------------------------------------------------------
Date: Wed, 1 Feb 89 16:37:13 PST
>From: emma@csli.Stanford.EDU (Emma Pease)
To: friends@csli.Stanford.EDU
Subject: CSLI Calendar, February 2, 4:14
C S L I C A L E N D A R O F P U B L I C E V E N T S
_____________________________________________________________________________
2 February 1989 Stanford Vol. 4, No. 14
_____________________________________________________________________________
A weekly publication of The Center for the Study of Language and
Information, Ventura Hall, Stanford University, Stanford, CA 94305
____________
LINGUISTICS DEPARTMENT COLLOQUIUM
The Mapping Between Phonological Categories and Phonetic Continua
Some Case Studies
Michel T. T. Jackson
Yale University
Friday, 3 February, 4:15
Cordura Conference Room
____________
LINGUISTICS DEPARTMENT COLLOQUIUM
The Expression of Arguments in Serial Verb Constructions
Mark Baker
McGill University
Tuesday, 7 February, 7:30
Cordura Conference Room
-----------
COMMONSENSE AND NONMONOTONIC REASONING SEMINAR
A Minimal Model Semantics with Default Priorities
Paul Morris
IntelliCorp
Monday, 6 February, 3:15pm
MJH 301
Existing default reasoning systems may be divided into minimality
based formalisms, such as circumscription, and those that depend on a
fixed point construction, like default logic. The fixed point schemes
have appeared to possess an advantage in allowing implicit
specification of arbitrary priorities among defaults. However, they
also have disadvantages, including a lack of cumulativity, and
difficulty in properly representing some situations where a mere
possibility of some contingency is sufficient to overcome a default.
We present a model minimization scheme that supports implicit
specification of priorities among defaults. The system enjoys
cumulativity (like other model preference systems), and gives more
satisfactory results in situations where a possibility overcomes a
default.
-----------
SYMBOLIC SYSTEMS FORUM
The Ecology of Computation
Bernard A. Huberman
Dynamics of Computation Group, Xerox PARC
Friday, 10 February, 3:15
Room 60:61G
A most advanced instance of concurrent computation is provided by
distributed processing in open systems that have no global controls.
These emerging heterogeneous networks are becoming self-regulating
entities, which in their behavior are very different from their
individual components. Their ability to remotely spawn processes in
other computers and servers of the system offers the possibility of
having a community of computational agents that, in their
interactions, are reminiscent of biological and social organizations.
This talk will give a perspective on computational ecologies, and
describe a theory of their behavior, which explicitly takes into
account incomplete knowledge and delayed information on the part of
its agents. When processes can choose among many possible strategies
while collaborating in the solution of computational tasks, the
dynamics leads to asymptotic regimes characterized by fixed points,
oscillations, and chaos. We will also describe Spawn, an ongoing
project that utilizes idle computational resources in a distributed
network of high-performance workstations. Finally, we will discuss
the possible existence of a universal law regulating the way in which
the benefit of cooperation is manifested in the system.
------------------------------
Date: Wed, 8 Feb 89 12:40:39 -0500
>From: finin@PRC.Unisys.COM
To: nl-kr@turing.cs.rpi.edu
Subject: The PRATTFALL Machine-Translation Module (Unisys AI Seminar)
AI SEMINAR
UNISYS PAOLI RESEARCH CENTER
Pundit's First French Lesson:
The PRATTFALL Machine-Translation Module
Francois-Michel Lang
Unisys Paoli Research Center
(lang@prc.unisys.com)
This talk describes PRATTFALL (Pundit Reads And Translates Texts into
French, ALbeit Loosely), a machine-translation module built on top of
the PUNDIT natural-language text-understanding system. Both PRATTFALL
and PUNDIT are implemented entirely in Quintus Prolog; PRATTFALL is
based on Michael McCord's LMT system.
There are five principal components in PRATTFALL, each of which is
described in turn:
* Transfer from English to French
* Restructuring to conform to French syntax
* Assembling lexical items
* Inflection of lexical items
* Cleanup of inflected lexical items
The transfer component is implemented using a definite-clause grammar
modelled after one of the components of PUNDIT's SPQR module.
Familiarity with LMT, PUNDIT, and machine translation in general is
helpful, but not necessary.
2:00 pm, Weds., February 15, 1989
Cafeteria Conference Room
Unisys Paoli Research Center
Route 252 and Central Ave.
Paoli PA 19311
-- non-Unisys visitors who are interested in attending should --
-- send email to finin@prc.unisys.com or call 215-648-7446 --
------------------------------
>From: Geert Adriaens <siegeert%BLEKUL60.BITNET@CUNYVM.CUNY.EDU>
Date: Tue, 31 Jan 89 14:02:59 +0100
To: nl-kr@cs.rochester.edu
Subject: MAIL-ADDRESSES
Can't postnews, so trying this way.
Desperately looking for mail-addresses of people whose last names are:
FANTY
MATSUMOTO
KAWAMOTO,
all active in NLP research.
Sorry if this screws up something.
Cheers,
Geert Adriaens
SIEMENS-METAL Project e-mail:
Maria Theresiastraat 21 siegeert@kulcs.uucp
B-3000 Leuven siegeert@blekul60.bitnet
tel: ..32 16 285091 siegeert@cs.kuleuven.ac.be
------------------------------
To: comp-ai-nlang-know-rep@rutgers.edu
>From: nurk@unix.eta.com (Tom Nurkkala)
Newsgroups: sci.lang,comp.ai.nlang-know-rep
Subject: Parsing Greek
Keywords: Greek, natural language, computational linguistics, parsing
Date: 9 Feb 89 21:44:03 GMT
Organization: ETA Systems Inc., St. Paul, Mn.
Does anyone out there have experience with, or references to, parsing
(ancient/koine) Greek text? I'm interested in developing tools for
detailed syntactical analysis of greek texts--any related information
would be welcomed.
Please mail to me...
--
Tom Nurkkala nurk@zwingli.unix.eta.com
Software Engineer
ETA Systems, Inc.
1450 Energy Park Drive "My brain is just a BUNDLE of nerves."
St. Paul, MN 55108
612/642-8390
------------------------------
End of NL-KR Digest
*******************