FROM TEXTS TO TAGS: ISSUES IN MULTILINGUAL LANGUAGE ANALYSIS
Workshop organized by the ACL special interest group SIGDAT
to be held in conjunction with the meeting of the European Chapter
of the Association of Computational Linguistics
Date: April 1, 1995
Place: Dublin, Ireland
Submission deadline: Jan 23
Notice of acceptance/rejection: February 10
Final papers due: March 1
The papers will be published as workshop notes.
With the growing amount of multilingual corpus data becoming
available, there is a pressing need to explore issues in
representation and analysis of these texts. Although extensive and
leading work has been accomplished for languages such as English, for
the most part many theoretical and concrete issues need to be resolved
in the representation and tagging of other languages.
The focus of this workshop is on multilingual text analysis, from the
level of text itself, e.g. tokenization, sentence separation, etc, to
morphosyntactic analysis, specifically tagging. We intend to focus on
tagging since it appears to be the case that, from a computational
point of view, part of speech tagging is often an important
prerequisite to further structural analysis. Additionally, many NLP
systems can make use of tagged corpora for various applications.
However, tasks such as tokenization and tagging continue to raise
serious challenges in multilingual text analysis, due to differing
types of morphological characteristics across languages.
Topics of interest include (but are not limited to):
- tokenization and segmentation
- interfaces between morphological analysis and part-of-speech tagging
- size and choice of tagset
- defining and refining new tag sets
- mapping between tag sets
- universal vs. language specific tags
- multilingual approaches to tagging
We invite submissions on topics that in general reflect an awareness
of differences and similarities in working on multilingual text.
We also welcome substantive descriptions
of newly started and ongoing projects.
Evelyne Tzoukermann Susan Armstrong-Warwick
AT&T Bell Laboratories ISSCO University of Geneva
Room 2D-448, P.O. Box 636 54 route des Acacias
600 Mountain Avenue
Murray Hill, NJ, 07944-0636 CH-1227 Geneve
tel. +1-908-582-2924 +41-22-705-7113
fax +1-908-582-7308 +41-22-300-1086
email [hidden email][hidden email]