===========================================
                SECOND
    CALL FOR PAPERS for the Workshop

    NLP for Under-Resourced Languages

          Friday, 10 June 2005

         in Dourdan, near Paris
============================================

Held in conjunction with the conference
TALN 2005 (6-10 June 2005)
http://www.limsi.fr/TALN05

Many languages have little or no information technology available:
they have no substantial presence on the Internet, and existing software 
has not been adapted for their use.  These are languages of countries in 
the developing world (e.g. Wolof in Senegal), or regional languages in 
countries where the first language is a global one (e.g. Breton in France).

Linguistic work on these languages is often lacking, and must overcome a 
number of difficulties:
- the presence of many lexical alternatives
- multiple spellings for the same word,
- the lack of exhaustive lexicons,
- non-standardized transcription methods
- etc.

This is why we describe these languages as "under-resourced" or
"pi-languages": for political and economic reasons they lack sufficient 
resources in terms of both linguistic study and information technology.

The goal of this workshop is to provide a progress report  on techniques 
being used for Natural Language Processing of under-resourced languages.

THEMES

Researchers are invited to present work  related to any of the following 
topics:

- Methods for measuring the level of technology available for a
  given language
- Assessment of resources for basic computerization, and standardization
  of these resources (editors, virtual keyboards, printing, sorting, etc.)
- Collection of corpora and other linguistic resources (lexicons,
  grammars); the role of the Internet, diasporas, unsupervised
  training algorithms
- Methodologies for computerization of under-resourced languages
  (details of techniques)
- Specific technologies (OCR, PDA)
- Encoding of linguistic data (lexicons, morphology, syntax, and
  semantics) and genericity of formats.
- Reports concerning the computerization of a particular language or
  group of pi-languages.
- Architectures for NLP technologies : adaptation of existing
  technologies to pi-languages.
- Surveys of major problems encountered in this area.

SELECTION CRITERIA
------------------------------
Authors are invited to submit papers concerning original  research.
All submissions will be examined by at least two specialists in the area.
Particular attention will be paid to:

- appropriateness for the themes of the workshop
- importance and originality of the contribution
- validity of the technical and scientific content
- critical discussion of the results, in particular as they
  relate to other work in the field
- organization and clarity of the presentation

The selected articles will be published in the conference proceedings.

METHODS OF SUBMISSION
------------------------------
* Submitted articles may not exceed 10 pages in Times 12,
  single-spaced (about 3000 words), figures, examples, and
  references included.
  Articles are to be written in French for French speakers and in
  English for non-French speakers.

* A LaTeX style file and Word template file will be available from
  the conference web site http://www.limsi.fr/TALN05.

* Articles must be submitted as an attachment to an email with the
  subject "atelier TALN soumission" and containing the title of the
  article, principal author's name, affiliation, postal address,
  email address, telephone number, and fax number.
  Send this email to chantal.enguehard@ univ-nantes.fr.

* It is IMPERATIVE that one of the following file formats be used,
  formatted for A4 paper: PS, PDF, RTF (Word).

* In case an email submission is not possible, paper submissions
  may be accepted.  Send 3 paper copies of the article to the
  following address:

Chantal Enguehard - TALN 2005
LINA
2, rue de la Houssinière
BP 92208
44322 Nantes Cedex 03
France

CALENDAR

Submission deadline: Thursday, 10 March 2005
Notification to authors: Tuesday, 5 April 2005
Final version: Friday, 15 April 2005
Conference : Friday, 10 June 2005

PROGRAM COMMITTEE

. Denis Béchet - Laboratoire d'Informatique Nantes-Atlantique
. Vincent Berment - Laboratoire Communication Langagière et Interaction 
Personne-Système
. Christian Boitet - Laboratoire Communication Langagière et Interaction 
Personne-Système
. Malek Boualem - France Telecom
. Marcel Diki-Kidiri - Langage, Langues et Cultures d'Afrique Noire
. Chantal Enguehard - Laboratoire d'Informatique Nantes-Atlantique (President)
. Laura Monceaux - Laboratoire d'Informatique Nantes-Atlantique 
. Kevin Scannell - Saint Louis University 
. Monique Slodzian - Centre de Recherche en Ingénierie Multilingue