Multilingual e Learning in LANGuage Engineering Project Overview

Multilingual e. Learning in LANGuage Engineering

Project Overview q Project span: Oct 2004 – Oct 2007 q Kick-off meeting Oct 9 -10 2004 q Project goals: q q q Produce language training resources Tailored to the needs of the translation market Deliverables in: CA, DE, EN, ES, FR, IT, (CZ)

Promoter Intercultural studies and Applied Languages Department, University Paris 7 Denis Diderot

Academic partners Ø Centre for Translation Studies, University of Leeds, UK Institute for Applied Linguistics, Translation and Interpreting, University of Saarland, Germany Ø Ø Department of Translation and Philology, University Pompeu Fabra, Barcelona, Spain Advanced School of Modern Languages for Interpreters and Translators, University of Bologna in Forli, Italy Ø Ø Translation and Interpreting School, University of Geneva, Switzerland

Academic partners Ø Institute of Translation and Interpreting, Milton Keynes, UK Ø Translation and Interpretation Institute, University of Ø Praetorius France, Gif-sur Yvette, France: localisation, Ø Olomouc Training Center, Olomouc, Czech Republic: Vienna, Austria translation, technical writing management and linguistic skills

Project Objective Adapt vocational training of translators and language professionals to the new needs of the market

Context q q Changes in the translation market New skills and competences required Ø Focus on innovative, dynamic, and collaborative learning environments

Target audiences q Students in initial vocational training q Practising translators and language professionals q Trainers of translators and language professionals

Target sectors q Education q q translation departments institutions focussing on training language professionals q Industry q q q translation market language industry other industries dealing with multilingual language resource management

Objectives q Propose a methodology for the collaborative creation of corpusbased e. Learning teaching content in translation q Design a framework for an European Master in translation

User needs evaluation q April 2005: questionnaire sent to q ITI’s 3000 members q translation lists in partner’s countries q Questions on q IT tools q Corpora use q e. Learning

Results: web vs. corpora q 600 responses in UK, France, Italy, and Germany q 90% professionals, 10% students q 90% use q q Google of these, 68% refine their searches 40% collect specific texts for translating so 40% use corpora

Responses: Corpus Tools n n 65. 9% Search facility in word processor 19. 0% Concordancer 14. 4% Other search tools (Trados, Concordance in translation memory) 0. 7% UNIX utilities

Responses: e. Learning domains of interest : q 28. 2% Exploitation of the web for linguistic purposes q 25. 7% IT skills for translators q 24. 1% Building their own corpora for specific projects q 19. 7% Exploitation of corpus data

Corpus-based approach q Comparable and parallel corpora q Learner Translator Corpus (LTC): q on-line submission of translations q students and professionals q Meta-data (L 1, L 2, training, etc. ) q First round of translation collection: EC document on workers’ mobility q Upcoming rounds : journalistic, administrative and technical translation

Error-annotation of the LTC q Short annotation scheme Content transfer errors q Language errors q q Translated in all partners’ languages q Used by teachers to annotate translations q XML annotation tool developed at the University of Geneva

LTC Design All translations are aligned with source texts q Linked to anonymised translator’s metadata q s. TT 1 c. TT 1 s. TT 2 SL 1 TL 1 s. TT. . . SL 2 TL 2 r. TT 1 SL… TL… Abbreviations T – Text S – Source L – Language T – Target s. TT - student Target Text c. TT - corrected student Target Text r. TT - reference Target Text

Online Content Topics q Machine translation q Translation memory q Terminology q Corpus use q Localization q Project management q Information management q Specialised q Mark-up translation languages

e. Learning Development Methodology Proposing a general methodology (‘best practices’) for collaborative online course creation in terms of: q Course design q Meta-data q e. Learning standards conformance q Implementation

Course development According to Bologna process hierarchy: Units-> Courses-> Modules (smallest to largest) q Content organized into units according to: q Learning outcomes (competencies) q Size of the unit (rough time estimate) q Objectives and Pedagogical approach q Resources and Prerequisites q Evaluation q Content / Activities q Relation to other units (reducing the number of credits needed) q

Learning Objects (LOs) q Building block for Units q Discussion on meta-data to be used (subset of LOM) q Los designed according to Moodle e. Learning platform activities: q q q Lessons Tests Quizzes Workshop Resources (e. g. LTC, parallel corpora)

e. Learning Platform q Moodle chosen for its sound pedagogical principles and features it offers q e. Learning standards compliance: q SCORM (interoperability, reusability) q LOM (meta-data description)

Course Testing Teaching material and online courses will be tested in: q 2006 with students, trainers and professionals q 2007 idem + CEE countries universities

Quality Assurance q Ensure that the contents are adapted to real market needs q Maintain contact with industrial development

European Masters § Existing curricula at partners’ sites § Existing curricula related to the field § Development within the consortium of a general framework for a European Masters (curriculum down to learning objects) § Complete or partial adoption byall partner institutions

Dissemination § Results tested with universities from CEE countries § Current results presented at conferences § Organisation of workshops § Dissemination to § translation companies § companies in need of multilingual language management