[Elsnet-list] Preliminary announcement of the Second Anaphora Resolution Exercise (ARE)

Constantin Orasan C.Orasan at wlv.ac.uk
Mon Nov 24 00:17:18 CET 2008


Preliminary announcement of the Second Anaphora Resolution Exercise
http://www.anaphora-and-coreference.info/ARE2009 

We are pleased to announce the second edition of the Anaphora Resolution
Exercise (ARE) organised in conjunction with the 7th Discourse Anaphora
and Anaphor Resolution Colloquium (DAARC2009), between 5 and 6 Nov 2009.
The general objective of ARE is to develop discourse anaphora resolution
methods and to evaluate them in a common and consistent manner. The
Anaphora Resolution Exercise (ARE) was started with the goal of
developing discourse anaphora resolution methods and evaluating them in
a common and consistent manner. 

ARE2009 will focus on pronominal and NP coreference for Catalan, Dutch,
English and Spanish. In addition to this, a multilingual personal
pronoun resolution task will be offered together with a several
preprocessing tasks which will focus on identification of
non-referential pronouns for Dutch and English and identification of
elliptical subjects in Catalan and Spanish. 

The exercise proposes four anaphora/coreference resolution tasks: 

1. Pronominal resolution on pre-annotated texts.
2. Coreferential Chains resolution on pre-annotated texts. 
3. Pronominal resolution on raw texts. 
4. Coreferential Chains resolution on raw texts. 

The preprocessing tasks proposed are: 

1. Identification of pleonastic "it" pronouns in English texts 
2. Identification of pleonastic "het" pronouns in Dutch 
3. Identification of elliptical subjects in Spanish and Catalan 

Multilingual pronoun resolution task: 

The goal of this task is to encourage participants to develop 
multilingual pronoun resolution systems by giving them paragraphs in 
Catalan, Dutch, English and Spanish with the referential pronouns 
marked. Participants will have to determine the antecedent of a 
pronoun. 

Real time evalution of systems: 

If enough interest is expressed by the participants, these tasks will be
also offered as real-time tasks at DAARC2009. In the real-time 
evaluation, participants will have to solve the chosen tasks in one 
hour. The texts given for processing will be similar to those in the 
training data, but they could also include some different texts in order
to test the scalability of the systems. 

Proceedings:

Participants will be encouraged to submit 4 page reports which will be 
published in the DAARC proceedings. The results of this evaluation will 
be disseminated at DAARC where we plan to organise a round table and/or 
a poster session on this theme. 

Tentative Timeline: 

24th Nov 2008: the first announcement of ARE 
15th Jan 2009: release of the training data 
4th - 23rd May 2009 - Test data is released 
30th May 2009 - Results communicated back to participants 
6 June 2009 - 4 page technical reports due from participants 
20 June 2009 - Reviews back to participants 
1 July 2009 - Final version of technical reports 
5 - 6 Nov 2009 - DAARC2009, Goa, India 

Participation is open to all. To indicate your intention to participate 
or for any other issues please contact us on 
ARE2009 at anaphora-and-coreference.info 

Web page: http://www.anaphora-and-coreference.info/ARE2009 

Organising committee: 
 * Antonio Branco, University of Lisbon, Portugal 
 * Dan Cristea, "Alexandru Ioan Cuza" University, Iasi, Romania 
 * Véronique Hoste, Hogeschool Gent, Belgium 
 * Ruslan Mitkov, University of Wolverhampton, UK 
 * Constantin Orasan, University of Wolverhampton, UK 
 * Marta Recasens, Universitat de Barcelona, Spain 
 * Sobha L, AU-KBC Research Centre, India

-----------
Dr. Constantin Orasan <C.Orasan at wlv.ac.uk>
Senior Lecturer in Computational Linguistics
Research Group in Computational Linguistics
http://www.wlv.ac.uk/~in6093/
University of Wolverhampton
-- 
Scanned by iCritical.


More information about the Elsnet-list mailing list