1st Workshop on Evaluation of Human Language Technologies for modern iberian languages (IBEREVAL) 
SEPLN 2010 satellite workshop

2010.07.21 Program
2010.06.16 DEADLINE EXTENDED TO JUNE 21, 2010.

With the aim of encouraging and promoting the development of Human Language Technologies (HLT) for modern iberian languages, the goal of this workshop is creating a discussion forum about different aspects regarding the evaluation of Natural Language Processing systems, emphasizing the main problems that modern iberian languages implied: Spanish, Catalan, Basque, Galician and Portuguese.

The implementation of this type of systems requires a deep evaluation and validation of them. In order to validate the HLT, similar evaluation methods must exists for similar applications, in order to compare them straightforwardly.

Besides, most tasks of HLT, as summarization and automatic translation, does not have a unique valid solution, but more than one can be accepted. Therefore, the evaluation methods required must be clear and adjusted to the system goals.

Following the line of other workshops focused in the evaluation of HLT systems, this IBEREVAL I workshop is presented as a discussion forum about the evaluation of systems working with modern iberian languages, with the aim of discovering the interests and necessities of the scientific community, dealing with the problems that evaluation implies, and promoting the creation of evaluation resources and training corpus for HLT systems working on these languages.


This IBEREVAL I workshop will bring together researchers, academic institutions and companies working on the development of HLT systems and their evaluation, participating through submissions regarding issues such as:

Development of evaluation resources over different HLT modules and applications:
  • creation of gold-standard corpus for iberian languages (Catalan, Basque, Galician, Portuguese and Spanish); Part of Speech tagging; syntactic parsing; semantic interpretation; anaphora resolution; information retrieval and extraction; summarization; machine translation, etc.
  • Evaluation metrics for the different HLT modules and applications.
  • General modules of evaluation: direct and indirect evaluation.
  • Development of testing, supervision and evaluation frameworks for HLT systems.
  • Manual, supervised and semi-supervised evaluation.
  • Creation and development of evaluation standards.
[EspaÃol] Espaņol
[English] English

ibereval10 [at]   |   Last updated: 09:10 21/07/2010 | Accesses since 20/04/2010: Comptador
XHTML 1.0 & CSS2 compliant.