LINGUIST List 14.1143

Fri Apr 18 2003

Calls: Computational Ling; Machine Translation

Editor for this issue: Marie Klopfenstein <marielinguistlist.org>

Directory

1. Menno van Zaanen, Workshop on Learning Context-Free Grammars
2. Leslie Barrett, Workshop on MT Evaluation

 First call for Papers and Participation:
 ECML/PKDD 2003 Workshop on Learning Context-Free Grammars
 (http://www.cs.kuleuven.ac.be/conference/ecmlpkdd/
 http://ilk.uvt.nl/~mvzaanen/ECMLPKDD/index.html)
 To be held at Dubrovnik, Monday, September 22, 2003

Most attention in the field of grammar induction has been set on the
problem of learning finite state automata, representing regular
languages. Yet long term dependencies, palindromic structures,
parenthesis are all internal structures that may appear in a wide
range of applications and are better described by context free
grammars. The purpose of the workshop is to provide a forum specific
to this question, enabling researchers to present their most recent
results over the issue of learning context-free grammars.

We encourage submissions related to the following topics:
* General learning results concerning the learnability of
 context-free grammars (or sub-classes of these), 
* Classes of context-free grammars for which positive learning
 results can be obtained, 
* Heuristics, and results on typical grammars, 
* Validation issues: benchmarks for context-free grammar learning, 
* Applications of context-free grammar learning, 
* Stochastic context-free grammar learning, 
* Learning context-free grammars from structured data, or
 semi-structured data, 
* Learning tree automata. 

Important dates:
* Workshop paper submission deadline: June 13, 2003. 
* Workshop paper acceptance notification: July 4, 2003. 
* Workshop paper camera-ready deadline: July 11, 2003. 
* Workshop to be held on the 22nd or 23rd of September, 2003. 

Submission Guidelines:
Submissions should be sent by June 13, 2003, in the electronic form as
a PDF or PostScript file to Menno van Zaanen (mvzaanenuvt.nl),
Subject: Grammatical Induction workshop submission paper. The papers
should be formatted as for the main ECML/PKDD-2003 conference
submissions. The maximum length of papers is 12 pages.
Submitted papers will be reviewed by referees from the Program
Committee. Accepted papers will be published in the working notes
provided by ECML/PKDD-2003. 

For a full description of the workshop focus and goals, visit the
website at
http://ilk.uvt.nl/~mvzaanen/ECMLPKDD/index.html
(also from the ECML/PKDD main page
http://www.cs.kuleuven.ac.be/conference/ecmlpkdd/).

Chairs: Colin de la Higuera, Pieter Adriaans, Menno van Zaanen.


- ----------------------------- Structural linguistics is a bitterly divided
- Menno van Zaanen - and unhappy discipline, and a large number of
- mvzaanenuvt.nl - practitioners spend too many nights drowning
- http://ilk.uvt.nl/~mvzaanen - their problems in Ouisghian Zodahs.
- ----------------------------- -Hitchhikers Guide to the Galaxy

Workshop on MT Evaluation

Towards Systematizing MT Evaluation

Theme:
Estimating the quality of any machine-translation system accurately is
only possible if the evaluation methodology is robust and
systematic. The NSF and EU-funded ISLE project has created a taxonomy
that relates situations and measures for a variety of MT
applications. The "Framework for MT Evaluation in ISLE" (FEMTI) is now
available online at
http://www.issco.unige.ch/projects/isle/taxonomy3/. The effort of
matching these measures correctly with their appropriate evaluation
tasks, however, is an area that needs further attention. For example,
what effect do "user needs" have on the "functionality
characteristics" specified in the FEMTI guidelines? To what extent are
there unseen relationships in the branches of the taxonomy? How can we
judge when a given evaluation measure is appropriate? Issues that come
to bear on this question are the automation of MT evaluation, the
extension to MT applications such as automated speech-translation, and
the evaluation of the very training corpora that an MT system relies
on to improve output quality.

This workshop welcomes papers for 30-minute presentations on the
comparison between MT evaluation measures, studies of the behavior of
individual measures (i.e., meta-evaluation), new uses for measures,
analysis of MT evaluation tasks with respect to measures, and related
topics on this theme. We solicit submissions to the workshop that
address some of the following issues, however any other topic related
to MT Testing and Evaluation is also acceptable:

Machine Translation Evaluation Measures:
 a.. Use of existing measures in the ISLE hierarchy (FEMTI
guidelines)
 b.. New measures and their uses
 c.. Matching evaluation requirements (e.g., translation tasks, user
profiles) with measures
 d.. Effects of combining measures 

Evaluation Measures and Languages
 a.. Is a metric's effectiveness language independent?
 b.. Counting grammatical features for evaluation

Evaluation and Domains
 a.. Measures for spoken Language translation
 b.. Domain-specific evaluation techniques
 c.. Using measures to evaluate the quality of a training corpus for
a given task

Automation vs. Human Testing
 a.. Which measures are suitable for automation?
 b.. Human/machine scoring comparisons
 c.. Human tester agreement: which measures fare best?

Submission Format

Papers (full papers up to 8 pages in length) must be submitted
electronically to: barrettsemanticdatasystems.com, or
andrei.popescu-belisissco.unige.ch

Papers are preferred in .pdf, .ps, .rtf, or .txt, format

Important Dates:

Paper submission deadline: May 11

Notification: June 30

Camera-ready due: July 31

Note: the workshop will be held on September 23, 2003

Also see workshop website for updates:

http://www.issco.unige.ch/projects/isle/MTE-at-MTS9.html

Contacts

Leslie Barrett (Transclick, Inc., New York, NY)

Andrei Popescu-Belis (ISSCO/TIM/ETI, University of Geneva)

Organizers

Leslie Barrett (Transclick, Inc., New York, NY)

Maghi King (ISSCO/TIM/ETI, University of Geneva)

Keith Miller (MITRE Corp)

Andrei Popescu-Belis (ISSCO/TIM/ETI, University of Geneva)

Program Committee:

Bonnie Dorr (University of Maryland)

Eduard Hovy (Information Sciences Institute, University of Southern
California)

Maghi King (ISSCO/TIM/ETI, University of Geneva)

Bente Maegaard (Center for Sprogteknologi, Copenhagen, Denmark)

Keith Miller (MITRE Corp.)

Martha Palmer (University of Pennsylvania)

Ted Petersen (Univesity of Minnesota)

Andrei Popescu-Belis (ISSCO/TIM/ETI, University of Geneva)

Florence Reeder (MITRE Corp)

Nancy Underwood (ISSCO/TIM/ETI, University of Geneva)

Michelle Vanni (National Computer Security Center)