Please accept our apologies for cross postings
2nd CALL FOR PAPERS
Lexicographic applications of computational approaches to collocations:
Restricted collocations in dictionaries
Co-Chairs: Geoffrey Williams (Université de Bretagne Sud, France), Ulrich Heid
(IMS Stuttgart, Germany)
This call for papers concerns a workshop dedicated to the theme of restricted collocation. This workshop is an integral part of the EURALEX 2002 conference and will form one of the parallel afternoon sessions.
Much work has been done over the years on all idiomatic aspects of language, this workshop proposes to deal with one that has attracted a lot of attention in computational and lexicographic circles, restricted collocation. Given that there is no generally accepted definition of this phenomenon, we accept this to cover a wide range of stable and restricted collocations including some idioms.
Restricted collocation presents a particular problem to the language learner. These constructions have often been ignored in the past; even to this day many language teachers are totally unaware of the phenomenon. Semantically transparent forms as convene a meeting, sharp rebuke may not present a difficulty in decoding, but their arbitrary nature is problematic for the learner. A problem for the language learner of limited linguistic competence, collocation also presents difficulties for the translator, especially in specialised texts.
Corpus linguistics has made a more thorough investigation of the phenomenon possible with the result that an increasing number of monolingual dictionaries hold collocational information. This has been true for English for some time and now to an increasing degree in other languages. At the same time bilingual dictionaries for a number of language combinations have started to appear.
Restricted collocation also presents a major challenge in machine translation. Early rule-based systems ignored idiomatic phenomena, with obvious results in terms of translation quality. This is no longer the case. An increasing interest in computational linguistics has lead to improved extraction routines and an interest in reusing material from paper dictionaries to create machine-readable databases.
Collocation has been the subject of a number of well attended workshops from computational and corpus linguistics aspects:
Marburg March 2000. Kollokationen, linguistiche Beschreibung und Akquisition aus Texten. Organised by Anke Lüdeling and Ulrich Heid, (IMS Stuttgart) and Petra Ludewig (University of Osnabruck), this workshop was held during the 22. Jahrestagung der Deutschen Gesellschaft für Sprachwissenschaft and brought together a number of linguists and computational linguists working on a variety of approaches to collocation covering a number of European languages, Dutch,
English, Finnish and German.
Paris. January 2001. Journée d'études de l'ATALA : La collocation. Organised by Béatrice Daille and Geoffrey Williams (Universities of Nantes and Lorient) brought together researchers in natural language processing working on French from a variety of approaches ranging from lexical functions to conceptual classes.
Toulouse. July 2001. Collocation: Computational Extraction, Analysis and Exploitation. Organised by Béatrice Daille and Geoffrey Williams (Universities of Nantes and Lorient), this workshop was held during the Association for Computational Linguistics conference and aimed to bring together computational linguists working on extraction and implementation of collocation.
This summer it is proposed to organise two closely related workshops; a first in Vienna focuses on more computational issues, the second would be held in conjunction with the Euralex Conference in Copenhagen.
Vienna. July 2002. This meeting on "Computational Approaches to Collocations" will be organised in Vienna by Brigitte Krenn (ÖFAI) in collaboration with Geoffrey Williams and Ulrich Heid. It will be chaired by Brigitte Krenn. The scope of the workshop will cover computational models and strategies for collocation identification and their use in computational linguistic applications. This includes statistics-based and hybrid methods on collocation identification, the development and testing of association measures, the discussion of significance versus relevance of identification results, the application of collocations in information extraction, in machine translation, in the development of lexical resources, in the evaluation of smoothing methods, etc.
After a series of workshops, which have covered a number of approaches and languages, the organisers believe that the time has come to link back into mainstream lexicography with a workshop held during the EURALEX conference in Copenhagen. The workshop would be chaired by Geoffrey Williams (Université de Bretagne Sud, Lorient) and Ulrich Heid (IMS Stuttgart).
The aim of this workshop is to look at the state-of-the-art in collocation from a lexicographic point of view, the advances made in paper and electronic dictionaries, both bilingual, monolingual, and their reusability for machine-readable dictionaries. By bringing together computational linguists, corpus linguists and lexicographers it is hoped that a fruitful exchange towards practical applications of collocation will be put into place. It is hoped that as wide a variety of languages as possible can be discussed as only through confronting the needs of a wide range of languages can the difficulties inherent in the concept of collocation be really grasped.
Submissions should follow the standard EURALEX format. Two hard copies and an electronic version in WORD.rtf are requested. These should be addressed directly to:
Dr Geoffrey Williams
Département Langues Etrangères Appliquées
U.F.R. Lettres et Sciences Humaines
4, rue Jean Zay
56321 LORIENT Cedex
The electronic version should be sent to Geoffrey.Williams@univ-ubs.fr with EURALEX submission as the subject.
The deadline for reception of submissions is: Monday 10th December 2001
This archive was generated by hypermail 2b29 : Sun Dec 02 2001 - 10:09:49 MET