Language Data Sharing in European Public Services – Overcoming Obstacles and Creating Sustainable Data Sharing Infrastructures

Proceedings of the 12th Conference on Language Resources and Evaluation (LREC 2020), pages 3443–3448 Marseille, 11–16 May 2020 c⃝ European Language Resources Association (ELRA), licensed under CC-BY-NC 3443 Language Data Sharing in European Public Services – Overcoming Obstacles and Creating Sustainable Data Sharing Infrastructures Lilli Smal1, Andrea Lösch1, Josef van Genabith1, Maria Giagkou2, Thierry Declerck1, Stephan Busemann1 1Deutsches Forschungszentrum für Künstliche Intelligenz (DFKI) GmbH, 2ATHENA Research and Innovation Center in Information, Communication and Knowledge Technologies 1Stuhlsatzenhausweg 3, 66123 Saarbrücken, Germany, 2Artemidos 6 & Epidavrou, 15125 Athens, Greece 1{lilli.smal, andrea.loesch, josef.van_genabith, thierry.declerck, stephan.busemann}@dfki.de, 2mgiagkou@athenarc.gr Abstract Data is key in training modern language technologies. In this paper, we summarise the findings of the first pan-European study on obstacles to sharing language data across 29 EU Member States and CEF-affiliated countries carried out under the ELRC White Paper action on Sustainable Language Data Sharing to Support Language Equality in Multilingual Europe. Why Language Data Matters. We present the methodology of the study, the obstacles identified and report on recommendations on how to overcome those. The obstacles are classified into (1) lack of appreciation of the value of language data, (2) structural challenges, (3) disposition towards CAT tools and lack of digital skills, (4) inadequate language data management practices, (5) limited access to outsourced translations, and (6) legal concerns. Recommendations are grouped into addressing the European/national policy level, and the organisational/institutional level. Keywords: ELRC, LR collection, language data sharing, language barriers, language technology, language equality, multilingualism, Europe 1. Introduction The European Union has 24 official languages. The European Parliament and the European Commission state that multilingualism is both “an asset and a shared commitment” (European Commission, 2008; European Parliament, 2009). Languages are not only a means of conveying information – they are constitutive parts of our cultures and identities, and thus one of the key pillars of Europe’s rich cultural heritage. At the same time, languages can also create barriers, and overcoming language barriers is one of the main challenges European citizens, public servants and businesses are facing in cross- language communication and trade. The European Commission Directorate-General for Translation has the largest translation service in the world translating EU legislation into all 24 official EU languages. In 2018, the Commission’s Directorate General for Translation translated some 2,255,000 pages (European Commission, 2019, p. 8). To meet the high demand for translation, the Commission developed its own machine translation (MT) system called eTranslation1 (formerly known as MT@EC), which was launched in 2013. In 2014, the European Commission started the Connecting Europe Facility (CEF) Programme. eTranslation aims to facilitate multilingual communication and the exchange of documents and other linguistic content in Europe between national public administrations on the one hand and between these administrations and EU and CEF-affiliated country citizens and businesses (European Commission, 2018) on the other hand. However, in order to make eTranslation work in the various domains and language pairs relevant to public 1 For more details on CEF eTranslation, see https://ec.europa.eu/cefdigital/wiki/display/CEFDIGITAL /eTranslation services and administrations across Europe, corresponding language resources were – …

Content truncated to 500 words