A Generator of Evaluative Arguments (GEA)
Giuseppe
Carenini
GEA is a fully-implemented, complete and modular NLG system for
generating user tailored evaluative arguments. GEA covers all aspects
of generating evaluative
arguments from selecting and organizing the content of the argument, to
expressing
the selected content into natural language. For content selection and
organization,
GEA applies an argumentation strategy based on guidelines from
argumentation
theory. For expressing the content into natural language, GEA relies on
a
set of techniques that extend and integrate previous work in
computational
linguistics on micro-planning and realizing evaluative arguments.
Finally,
a quantitative model of user preferences expressed as an additive
multiattribute
value function (AMVF) is the key knowledge source used by GEA in
tailoring
the content, organization and phrasing of the generated arguments to
its
users.
Software:
GEA is implemented in Allegro
Common Lisp. It runs on Sun
machines (Allegro 4.3.1 and Allegro 5.0).
I should be easy to port it to Allegro for Windows. If interested in obtaining the source code,
please contact carenini@cs.ubc.ca
Published papers on GEA:
- Giuseppe Carenini and Johanna Moore, A strategy for
generating Evaluative Arguments. International Natural Language
Generation Conference, Mitzpe Ramon, Israel.
2000.
- Carenini G., GEA: a Complete, Modular System for
Generating Evaluative Arguments. International workshop on
Computational Models of Natural Language Argument
San Francisco, California (USA), 2001.
Projects
that are extending and applying the GEA approach:
- AT&T
MATCH system
- A. Stent, M.
Walker, S. Whittaker and P. Maloor, "User-tailored generation for
spoken dialogue: An experiment", in Proceedings
of ICSLP 2002, September 2002.
- M. Johnston, S.
Bangalore, G. Vasireddy, A.
Stent, P. Ehlen, M. Walker, S. Whittaker and P. Maloor, "MATCH: An
architecture for multimodal dialogue systems", in Proceedings of
ACL 2002, July 2002.
- M. Walker, S.
Whittaker, A. Stent and P. Maloor, "Generating evaluative responses in
spoken dialogue", in Proceedings of INLG 2002, July 2002.
- Whittaker, S.,
Walker, M., and Moore, J.. Fish
or Fowl: A Wizard of Oz Evaluation of Dialogue Strategies in the
Restaurant Domain .Language Resources and
Evaluation Conference. 2002.
<> - <>FLIGHTS
system - School of Informatics (University of Edinburgh)
- <>Johanna D. Moore, Mary Ellen Foster, Oliver Lemon, and
Michael White,
Generating Tailored, Comparative Descriptions in Spoken Dialogue, in
Proceedings of the Seventeenth International Florida Artificial
Intelligence Research Sociey Conference, AAAI Press, 2004.
Please, send comments and inquiries to carenini@cs.ubc.ca