APPEL: revue TAL : Automatic Text Summarization: Solutions and Perspectives

Fri Jan 16 17:12:44 UTC 2004

Automatic Text Summarization: Solutions and Perspectives

Deadline: February 10th, 2004

Special Issue Editor: Jean-Luc Minel (LaLICC, CNRS)

Context: Automatic Text Summarization

 From a scientific point of view, the problem of summarization extends
beyond its immediate boundaries, largely due to certain rhetorical
relations that need to be taken into consideration, such as discourse
frames, or direct or indirect levels of discourse. In recent years,
this research area has moved beyond the processing of phrases as the
unit of analysis to include concept search and the design of suitable
methods for detecting and representing textual structures.

Until the mid-1990?s, scientific texts provided a field of
experimentation for automatic summarization methods, but the
digitization of texts and their increased availability on the Web and
intranets has fundamentally changed user needs and uses. The abundance
of texts brings with it an urgent requirement for the creation and
production of summarization tools capable of finding, selecting and
extracting textual information concisely.

In terms of the need, it has become necessary to bring web-search
tools and company networks together with tools for automatic
summarization.

In terms of use, the development of text-search tools requires careful
consideration of text representation and the design of user
interfaces, which in turn leads to studies being carried out in the
domain of information science on new types of written text.

Topic

This special issue aims:

- on the one hand, to present new approaches or methods which may lead
to promising prototypes for automated systems. It also concerns the
development of an awareness of the importance of automatic
summarization, alongside technology, for linguistic
engineering. Whatever the automatic summarization project, there are
only two underlying techniques for its realization. Firstly,
techniques which extract full phrases from source texts, and secondly,
those which generate a new, condensed text. Only the first kind of
technique has allowed for the implementation of systems that can be
considered as providing reasonable results according when evaluated
for comercial potential. The second technique is of interest in a
research context because of the various linguistic problems left
unsolved at the level of interpretation, in view of the limitations in
their computational implementation.

- on the other hand, to propose bridges between the various areas in
which text constitutes the main object of study (i.e., in the domain
of information science).

Papers are invited which contribute to the following themes:

- Numerical approaches versus linguistic approaches, with a particular
focus on papers that explore the complementarity between the two.

- The automatic detection of textual structures, including:

o the identification of topic
o the identification of argumentative structures
o the improvement of coherence and cohesion (dealing with anaphora, etc)

- Different dimensions of summary

o Types of summary
o Translingual summaries
o Multi-document summaries

- Linguistic resources necessary for summarization systems

o the application of terminology and ontologies
o Generic versus specific-purpose resources

- Summary and Normalization

o Integrating summary systems into networks
o Integrating summarization systems into linguistic (s?)

- Methods of assessment  for summarization systems
- Extension of the summarization issue to the semantic filtering of
   texts:

o Exploring new opportunities
o The production of summaries in response to specific user needs
o The use of navigation tools and interfaces which exploit textual
   structure
o summarization and annotation of texts in a Semantic Web context

Editorial Committee

John Atkinson  (Université de Concepcion, Chili)
Michel Charolles (LATTICE, Université Paris 3, France)
Jean-Pierre Desclés (LaLICC, Université Paris-Sorbonne, France)
Michael Elhadad (Computer Science Department, Ben Gurion University, Israel)
Noemie Elhadad (Computer Science Department, Columbia University, USA)
Guy Lapalme (Université de Montréal, Canada)
Inderjeet Mani (Société MITRE, USA)
Jean-Guy Meunier (UQUAM, Canada)
Dragomir Radev (University of Michigan, USA)
Antoinette Renouf (University of Liverpool, UK)
Horacio Saggion (Computer Science Department, University of Sheffield, UK)
Dina Wonsever (Université de la République, Uruguay)

Format

Contributions (25 pages maximum) should be submitted in Word,
Postscript or Acrobat format. The file styles are provided as part of
the regulations on the journal homepage.

Language

Papers should be written either in French or English. However, English
submissions will only be accepted from non-French speakers.

Dates

Submission deadline: February 10th, 2004
Final committee decisions: April 15, 2004.
The camera-ready version of the accepted articles should reach the
editors by June 1st, 2004, for publication in 2004.

Those who intend to submit an article are encouraged to contact
Dr. Jean-Luc Minel (jean-luc.minel at paris4.sorbonne.fr).

Paper submission

The articles should be submitted by electronic mail to:
jean-luc.minel at paris4.sorbonne.fr or by normal mail to the following
address:

Jean-Luc MINEL
Laboratoire LaLICC, UMR 8139 (CNRS - Université Paris-Sorbonne)
ISHA, 96 Boulevard Raspail
75 006 Paris

-------------------------------------------------------------------------
Message diffusé par la liste Langage Naturel <LN at cines.fr>
Informations, abonnement : http://www.biomath.jussieu.fr/LN/LN-F/
English version          : http://www.biomath.jussieu.fr/LN/LN/
Archives                 : http://listserv.linguistlist.org/archives/ln.html

La liste LN est parrainée par l'ATALA (Association pour le Traitement
Automatique des Langues)
Information et adhésion  : http://www.atala.org/
-------------------------------------------------------------------------