NLG workshop at KI99, Bonn Germany
becker at dfki.de
Fri Mar 19 12:27:46 EST 1999
First Call for Papers and Participation
``MAY I SPEAK FREELY?''
BETWEEN TEMPLATES AND FREE CHOICE IN NATURAL LANGUAGE
GENERATION: WHAT IS THE RIGHT NLG TECHNOLOGY FOR MY APPLICATION?
A workshop to be held at the German Annual Conference on Artificial
Intelligence (KI-99) in Bonn, Germany, September 13-15, 1999
A workshop of the Special Interest Group 1.3.1 on Natural Language
Systems of the Gesellschaft fuer Informatik e.V.
Organized by Tilman Becker and Stephan Busemann, DFKI
Deadline: May 15, 1999
AIMS AND SCOPE
This workshop will be an excellent opportunity to get an overview over
existing state-of-the-art technology in natural language generation
(NLG) and its optimal usage. The workshop will be relevant for both
developers and users of NLG systems. Exploring conditions for
successful NLG applications is a step that should be taken jointly by
technology providers and current and potential users of NLG software.
NLG technology is currently finding its way into commercial systems.
Promising applications are the automatic generation of weather
forecasts, information about measurement data, or various kinds of
authoring systems that help an author in composing a text.
Generally the techniques used in NLG applications and application-
oriented NLG systems differ from those utilized in research systems.
While the latter typically aim at general solutions, the former are
geared towards solving particular classes of NLG problems. This involves
dealing with canned texts or templates rather than with free choice from
the coverage of complex linguistic grammars. Correspondingly, different
systems encode linguistic knowledge at different levels of detail and
While many theoretical linguistic concepts are difficult to use in
practice, the techniques used in application systems often lack
theoretical foundation. Their linguistic inadequacy severely restricts
the transportability of systems to other tasks and domains. Current
work in NLG addresses this problem, e.g., the RAGS project
Whatever the chosen approach, its adequacy depends on the expected input
to, and the desired output of, the NLG system. Input may vary from
non-linguistic data to surface-semantic sentence representations.
Further generation parameters may be available, such as a user model or
a discourse model. Output may vary from a single sentence per context to
multiple alternatives that allow (require) the user to choose from.
This workshop aims at exploring the tension between more general and
more specific approaches to NLG, thereby clarifying what NLG technology
is suited best for which task. We want to discuss how different demands
on the input to a NLG system relate to different variability in the
output. Furthermore, the kind of input that is available from
applications using NLG modules and the expectations on such applications
will be a topic of interest.
We solicit contributions by researchers and developers working in the
field of NLG as well as by current and potential appliers of NLG
software. Contributions should belong to one of the following
- Presentation (or even better demonstration) of implemented systems
with an emphasis on the workshop's topic, as outlined above.
- Comparison of implemented systems regarding these issues and/or
attempts towards a cross-classification of techniques and tasks.
- Application demands placed on NLG modules with a description of the
potential output of the application and/or the expected output from
the NLG module.
Participants are invited to prepare short (15 minutes) or long (30
minutes) presentations, by submitting a two or four page abstract,
respectively. Submissions should be in the Springer LNCS-Format (see
submission of PostScript files is strongly preferred. Abstracts
submitted electronically will be made available in advance so that
participants can prepare for fruitful discussions following the
Submissions may be made in German or English; since we expect a number
of international participants, all presentations should be given in
Submissions must reach the organizers by May 15, 1999. Notification of
contributions will be given at the end of May. Accepted contributions
will be published in a volume complementing the proceedings of the KI-99
Note: The participation in the workshop requires the registration for
one of the supporting conferences, i.e., KI-99
(http://wob.informatik.uni-bonn.de/Wob/de/view/ki99.html) or DAGM '99
- Submission of contributions: May 15, 1999
- Notification of acceptance: May 31st, 1999
- Camera-ready copy due: June 16th, 1999
- KI-99 and Workshop: September 13-15, 1999
Dr. Tilman Becker and Dr. Stephan Busemann
(+49) 681-302-5271 and -5286 (phone)
(+49) 681-302-5341 and -5338 (fax)
becker at dfki.de and busemann at dfki.de
More information about the HPSG-L