Parallel Text Processing: Alignment and Use of Translation Corpora

416 Pages·2000·14.87 MB·English

by  Jean Véronis (auth.)

Parallel Text Processing
Text, Speech and Language Technology
VOLUME 13
Series Editors
Nancy Ide, Vassar College, New York
Jean Veronis, Universite de Provence and CNRS, France
Editorial Board
Harald Baayen, Max Planck Institute for Psycholinguistics, The Netherlands
Kenneth W. Church, AT & T Bell Labs, New Jersey, USA
Judith Klavans, Columbia University, New York, USA
David T. Barnard, University of Regina, Canada
Dan Tufis, Romanian Academy of Sciences, Romania
Joaquim Llisterri, Universitat Autonoma de Barcelona, Spain
Stig Johansson, University of Oslo, Norway
Joseph Mariani, UMSI-CNRS, France
The titles published in this series are listed at the end of this volume.
Parallel Text Processing
Alignment and Use of Translation Corpora
Edited by
Jean Veronis
Universite de Provence,
Aix-en-Provence, France
SPRINGER-SCIENCE+BUSINESS MEDIA, B.V.
A CLP. Catalogue record for this book is available from the Library of Congress.
ISBN 978-90-481-5555-2 ISBN 978-94-017-2535-4 (eBook)
DOI 10.1007/978-94-017-2535-4
Printed on acid-free paper
All Rights Reserved
© 2000 Springer Science+Business Media Oordrecht
Originally published by Kluwer Academic Publishers in 2000
No part of the material protected by this copyright notice may be reproduced or
utilized in any form or by any means, electronic or mechanical,
including photocopying, recording or by any information storage and
retrieval system, without written permission from the copyright owner.
But the LORD came down to see the Or Yahw? descendit pour voir fa ville et
city and the tower that the men were la tour que les hommes avaient bdties.
building.
The LORD said, "If as one people Et Yahve dit : « Voici que taus font un
speaking the same language they seul peuple et parient une seule lan
have begun to do this, then nothing gue, et tel est Ie debut de leurs entre
they plan to do will be impossible for prises ! Maintenant, aucun dessein ne
them. sera irrealisable pour eux.
Come, let us go down and confuse Allons! Descendons! Et ta, confon
their language so they will not dons leur langage pour qu 'ils ne
understand each other. " s 'entendent plus les uns les autres. »
So the LORD scattered them from Yahve les dispersa de la sur tOUfe la
there over all the earth, and they face de la terre et ils cesserent de bdtir
stopped building the city. la ville.
Genesis 11 :9, The Holy Bible: New Genese 11 :9, La Bible de Jerusalem
International Version
Contents
Foreword ........................................................................................................... xi
Terminological note .......................................................................................... xiii
Preface by Martin Kay ....................................................................................... xv
Contributors ..................................................................................................... xxi
INTRODUCTION
1. From the Rosetta stone to the information society ................................. 1
A survey of parallel text processing
Jean Veron is
ALIGNMENT METHODOLOGY
2. Pattern recognition for mapping bitext correspondence ....................... 25
1. Dan Melamed
3. Multilingual text alignment .................................................................. 49
Aligning three or more versions of a text
Michel Simard
V1l1 Contents
4. A comprehensive bilingual word alignment system ............................. 69
Application to disparate languages: Hebrew and English
Yaacov Choueka, Ehud S. Conley and Ido Dagan
5. A knowledge-lite approach to word alignment .................................... 97
Lars Ahrenberg, Mikael Andersson and Magnus Merkel
6. From sentences to words and clauses ................................................. 117
Stelios Piperidis, Harris Papageorgiou and Sotiris Boutsis
7. Bracketing and aligning words and constituents in parallel text using
Stochastic Inversion Transduction Grammars .................................... 139
Dekai Wu
8. The translation network ...................................................................... 169
A model for a fine-grained description of translations
Diana Santos
9. Parallel text alignment using cross lingual information retrieval
techniques ........................................................................................... 187
Christian Fluhr, Frederique Bisson and Faiza Elkateb
10. Parallel alignment of structured documents ....................................... 201
Laurent Romary and Patrice Bonhomme
ApPLICATIONS
11. A statistical view on bilingual lexicon extraction .............................. 219
From parallel corpora to non-parallel corpora
Pascale Fung
12. Terminology extraction from parallel technical texts ......................... 237
Ingeborg Blank
13. Term alignment in use ........................................................................ 253
Machine-aided human translation
Eric Gaussier, David Hull and Salah Ait-Mokhtar
14. Automatic dictionary extraction for cross-language information
retrieval ............................................................................................... 275
RalfD . Brown, Jaime G. Carbonell and Yiming Yang
Contents IX
15. Parallel texts in computer-assisted language learning ........................ 299
John Nerbonne
RESOURCES AND EVALUATION
16. Japanese-English aligned bilingual corpora ....................................... 313
Hitoshi Isahara and Masahiko Haruno
17. Building a parallel corpus of English/Panjabi .................................... 335
Sukhdave Singh, Tony McEnery and Paul Baker
18. Sharing oftranslation memory databases derived from aligned
parallel text ............................................. '" ......................................... 347
Alan K. Melby
19. Evaluation of parallel text alignment systems .................................... 369
The ARCADE project
Jean Veronis and Philippe Langlais
Index of terms .................................................................................................. 389
Index of authors ............................................................................................... 395
Index of languages and writing systems .......................................................... 401
Foreword
This book evolved from the ARCADE evaluation exercisel that started in 1995.
The project's goal is to evaluate alignment systems for parallel texts, i.e., texts
accompanied by their translation. Thirteen teams from various places around the
world have participated so far and for the first time, some ten to fifteen years
after the first alignment techniques were designed, the community has been able
to get a clear picture of the behaviour of alignment systems. Several chapters in
this book describe the details of competing systems, and the last chapter is
devoted to the description of the evaluation protocol and results. The remaining
chapters were especially commissioned from researchers who have been major
figures in the field in recent years, in an attempt to address a wide range of
topics that describe the state of the art in parallel text processing and use.
As I recalled in the introduction, the Rosetta stone won eternal fame as the
prototype of parallel texts, but such texts are probably almost as old as the
invention of writing. Nowadays, parallel texts are electronic, and they are be
coming an increasingly important resource for building the natural language
processing tools needed in the "multilingual information society" that is cur
rently emerging at an incredible speed. Applications are numerous, and they are
expanding every day: multilingual lexicography and terminology, machine and
human translation, cross-language information retrieval, language learning, etc.
After the introduction, which gives an overview of the field and places the
various individual chapters in context (Chapter 1), the book is divided into three
parts: alignment methodology (Chapters 2 to 10), applications (Chapters 11 to
IS), and resources and evaluation (Chapters 13 to 19). This division was made
for clarity's sake, and it is true that all chapters have a main focus that falls into
one of these three areas. However, the reader should keep in mind that in many
I http://www.up.univ-mrs.fr/-veronislarcade
XlI Foreword
cases, the individual chapters do address several topics. For example, a chapter
focussing on a given application may very well present an improvement in the
alignment techniques, and describe the corpus resources used or developed for
its training and evaluation of that application.
I would like to thank the thirty-four authors who contributed to the book, not
only for their outstanding papers, but also for the remarkable collective work
they have accomplished reviewing colleagues' chapters, helping with refer
ences, etc. I have caused them a great deal of trouble with deadlines and for
mats, and I hope they will forgive me. I would also like to express my gratitude
to the many people who read the various drafts of the book and provided their
very helpful remarks and comments, especially Ken Church and Stig Johansson,
as well ,~s the anonymous reviewers who thoroughly read all chapters and
helped "harmonize" the volume. Of course, as usual, remaining errors are our
own-and especially mine, since as editor I should have caught them. Thanks
go also to Joseph Mariani, who encouraged and supported the ARCADE project
which was the starting point of this collection.
I extend my warmest thanks to Martin Kay, who agreed to write a preface to
this book. This is a great honour for us, because Martin is not only one of the
fathers of computational linguistics, but also the person who, with his student
Martin R6scheisen, designed the first parallel text alignment system in 1987. In
his preface, Martin Kay modestly refers to that pioneering work as an exercise
with "nothing of importance" in its outcome, but we all now know that as early
as 1980, he had already realised how important translation corpora were, as
witnessed by his famous memorandum, The proper place of men and machines
in translation. His work on alignment was therefore hardly an accident, but
rather, reveals one of the many insights that Martin has had on what was to
become important for the future of computational linguistics. In a few simple
words here, Martin highlights the challenges that are facing us in the alignment
of finer-grained units such as words and clauses, and reminds us in a striking
way that (good) translation is not as simple and compositional an activity as we
like to pretend in trying to make our systems work.
I am convinced that the future will show that parallel corpora are modern
day Rosetta stones which have helped us better understand translation, that
incredibly complex activity, and hopefully, will help us do it some day as well
on machines as humans do---or at least not worse.
Jean Veronis, Aix-en-Provence, March 2000