Bente Maegaard

ELRA – European Language Resources Association-Background, Recent Developments and Future Perspectives

The European Language Resources Association (ELRA) was founded in 1995 with the mission of providing language resources (LR) to European research institutions and companies. In this paper we describe the background, the mission and the major activities since then.

PaTrans: a patent translation system

This paper describes PaTrans-a fully automatic production MT system designed for producing raw translations of patent texts from English into Danish. First we describe the backbone of the system: the EUROTRA research project and prototype. Then we give an overview of the translation process and the basic functionality of PaTrans, and finally we describe some recent extensions for improving processing efficiency and the translation quality of unexpected input encountered in real-life texts.

Eurotra: General system design

This article gives a general overview of the technical, scientific and philosophical character of the Eurotra machine translation program, outlines its developmental history and lists some plans for future work.

The Eurotra MT formalism

This paper gives a presentation of the E-framework, the formalism used in Eurotra for implementing the prototype translation system. In Section 1, we outline some of the underlying design principles and briefly discuss the evolutionary history of the Eurotra formalism, relating it to its predecessors and other recent unification-based formalisms. Section 2 introduces the reader to some central concepts and gives a comprehensive description of the core of the formalism.

The transfer of finite verb forms in a machine translation system

This paper is based on work done jointly by Hanne Ruus, Ebbe Spang-Hanssen and the author, all of the University of Copenhagen. The work was done within the framework of EUROTRA and sponsored by the Commission of the European Community. Work on this paper was begun whilst the author was at ISSCO, Geneva.

Observatory for Language Resources and Machine Translation in Europe – LT_Observatory

The European Digital Single Market, one of the main goals of Europe 2020, is still fragmented due to language barriers. Language technologies (LT), like Machine Translation (MT) solutions, are key elements for solving this fragmentation. Nevertheless, it is necessary to compile, benchmark the quality and facilitate the access to Language Resources to build successful MT solutions. With these aims, the LT_Observatory project has been developed (2014–2016). The project was funded by the European Commission through the H2020 programme. This article describes the main outputs: An on-line catalogue of language resources in existing pools and other national resources based on pre-identified user needs. Methodologies for improving the quality and usability of language resources. National and regional language strategies, policies and funding sources to support language technologies. An EcoGuide that aims to adapt the findings of the LT_Observatory project for various stakeholder groups providing practical information for operational usability of LRs and tools for MT application, funding opportunities, and recommendations geared at European, national and regional policy and decision makers.

Towards networking efforts to build an Arabic Cooperation Roadmap

The development of language resources and tools for the Arabic language is important for the economy in the Arab countries; it is also important for the culture. By focussing on Arabic language technology and making both the technology and content available in Arabic, the use of Arabic will grow and the demand for foreign language information will decrease. At the same time language technology can help access information in foreign languages, and finally, it can help spread Arabic ideas and culture to non-Arabic languages. The goal of the MEDAR project1, supported by the European Commission ICT2 programme, is to establish a network of partner centres of best practice in Arabic dedicated to promoting Arabic HLT3. MEDAR partners have collected knowledge about existing language resources, players, products etc, and based on this knowledge and additional research, the present report has been compiled. We will present a first version of the Cooperation Roadmap for Arabic language technology.

Evaluation of NLP systems

Computational linguistics as a science has had its evaluation methods since its early days: A concordance program can be evaluated according to its ability to find all occurrences, to list them properly, to have a flexible user interface etc., frequency programs *nay be evaluated according to their statistics, the possibility of lemmatisation, parsers are evaluated according to their efficiency etc. When we contemplate one component at a time and want a technical evaluation, we normally have no problem defining the evaluation criteria.

Designing and testing linguistic development phases in a machine translation project

In the first sections of the paper the design of a scheme for building up linguistic coverage in a multi-lingual machine translation project (EUROTRA) is discussed and a solution is proposed which also takes into account the extensibility of the system, and in the last section the aspect of testing is briefly discussed.