|
|
Natural Language based Knowledge Representations:
New Perspectives
Adam's
Mark Hotel
Clearwater Beach, FL
May 15-17, 2005
Paper submission
deadline: Friday, October 22, 2004.
Notifications
sent by: Wednesday, January 7, 2005.
Final papers
due: Friday, February 4, 2005.
Special
Track Coordinator
Vasile
Rus
Department of Computer Science The University of Memphis
Call for Papers
Goal
We define Natural Language based Knowledge Representation as a
representation that is closer to NL as opposed to being more artificial.
The intuition is that such representations will be more expressive, or
at least borrow as much as possible from the expressiveness of NL.
The goal of this Special Track is to re-assess the status of Natural
Language (NL) based Knowledge Representations (KR) and systems. It was
believed that NL-based KR systems would deliver representational and inferential
properties of natural language but the hard issues in NL such as ambiguity,
context-dependency and the complexity of syntax, semantics and pragmatics
limited in the past the progress of building promising knowledge processing
systems.
Among the advantages of building NL-based KR systems are:
-
NL-based systems would be user friendly
-
Most human knowledge is encoded and transmitted via natural language and
thus NL-based KR are a natural development
-
searching on the Internet has become a necessity and a daily task for most
of us; natural language is heavily used in this task since more than 90%
of the web information is textual
-
NL-based knowledge processing sytems would provide a uniform symbolic representation
for encoding knowledge and processing it
-
it is hard to match expressiveness and precision of natural language, particularly
in not (well) formalized domains
Recent advances of Natural Language Processing (NLP) in the areas of syntactic
parsing, semantics and pragmatics have opened new perspectives for developing
expressive KR and building promising NL-based knowledge processing systems.
A special track to re-assess the new perspectives is needed and this Special
Track aims to satisfy this need.
Topics
We invite highly original papers that describe:
-
novel, expressive NL-based representations
-
multi-level representations
-
NL-based inference methods and reasoning engines
-
evaluation techniques for NL-based KR
-
challenges in NL-based representations and in deriving such representations
from NL texts, especially how recent advances in NL technologies provide
new opportunities for knowledge aquisition and processing
-
techniques for coreference resolution, word sense disambiguation, information
extraction, predicate-argument structure, frame semantics, preposition
semantics, syntactic parsing, named entity recognition, etc. and their
impact on NL-based KR
-
NL-based KR in dialogue management
-
NL-based KR in Question/Query Answering
-
NL-based KR in auto tutoring systems
-
NL-based KR in Information Retrieval
-
large knowledge bases construction using NL-based KR
-
scalability issues of systems built using NL-based KR
-
other related issues
Submission Guidelines
Interested authors should format their papers according to AAAI formatting
guidelines. The papers should not exceed 6 pages and are due by October
22, 2004. Please note the change from 5 to 6 pages from the first CFP.
Additional pages (7 and more) have to be cleared by the program chairs
and will be $100 each. The papers should not identify the author(s) in
any manner. Authors should indicate the special track if one exists that
closely matches the topic of their paper. All submissions will be done
electronically via the FLAIRS web submission system available through the
paper submission site at http://earth.cs.ccsu.edu/~flairs/submission.html.
Conference Proceedings
Papers will be refereed and all accepted papers will appear in the conference
proceedings which will be published by AAAI Press. Selected authors
will be invited to submit extended versions of their papers to a special
issue of the International Journal on Artificial Intelligence Tools (IJAIT)
to be published in 2006.
Organizing Committee
Vasile Rus, The University of Memphis
Vivi Nastase, University of Ottawa
Programme Committee
Jerry Hobbs, USC/ISI
Andrew Gordon, USC/ICT
Art Graesser, University of Memphis
Bob Givan, Purdue University
Lucja Iwanska, Georgia Southwestern State University
Fernando Gomez, University of Central Florida
Susan Haller, University of Wisconsin - Parkside
Tudor Muresan, Technical University of Cluj
Stephen Anthony, University of Sydney
Rada Mihalcea, University of North Texas
Smaranda Muresan, Columbia University
Diana Inkpen, University of Ottawa
Zdravko Markov, Central Connecticut State University
Vasile Rus, The University of Memphis
Vivi Nastase, University of Ottawa
Boris Galitsky, University of London
David Ahn, University of Amsterdam
Valentin Jijkoun, University of Amsterdam
Max Louwerse, University of Memphis
Paul Morarescu, University of Texas at Dallas
Andrew Olney, Fedex Institute of Technology
Zygmunt Vetulani, Adam Mickiewicz University
Further Information
Questions regarding the NL-based
KR track should be addressed to the track co-chairs:
Vasile Rus at vrus@memphis.edu
Viviana Nastase at vnastase@csi.uottawa.ca
Questions regarding paper submission
should be addressed to the FLAIRS-2005 program co-chairs:
Ingrid Russell, irussell@hartford.edu, University of Hartford
Zdravko Markov, markovz@ccsu.edu, Central Connecticut State University
General questions concerning the
conference should be addressed to the FLAIRS-2005 conference co-chairs:
Diane Cook, University of Texas at Arlington
Lawrence Holder, University of Texas at Arlington
Special Tracks Coordinator
Todd Neller, Gettysburg College
Invited Speakers
Lawrence Hunter, University of Colorado
Martha Pollack, University of Michigan
Ted Senator, DARPA
David Stork, Ricoh and Stanford University
Conference Web Sites
Paper submission site:
http://earth.cs.ccsu.edu/~flairs/submission.html
NL-based KR special track web
page: http://www.cs.memphis.edu/~vrus/flairs05.html
FLAIRS-2005 conference web page:
http://www.flairs.com
Florida AI Research Society (FLAIRS):
http://www.flairs.com
|