ELRC Action: Covering Confidentiality, Correctness and Cross-linguality

Proceedings of the 13th Conference on Language Resources and Evaluation (LREC 2022), pages 6240–6249 Marseille, 20-25 June 2022 © European Language Resources Association (ELRA), licensed under CC-BY-NC-4.0 6240 ELRC Action: Covering Confidentiality, Correctness and Cross-linguality Tom Vanallemeersch,1 Arne Defauw,1 Sara Szoc,1 Alina Kramchaninova,1 Joachim Van den Bogaert,1 Andrea L¨osch2 1CrossLang Kerkstraat 106, 9050 Gent, Belgium {firstname.lastname}@crosslang.com 2DFKI Stuhlsatzenhausweg 3, D-66123 Saarbr¨ucken, Germany andrea.loesch@dfki.de Abstract We describe the language technology (LT) assessments carried out in the ELRC action (European Language Resource Coordination) of the European Commission, which aims towards minimising language barriers across the EU. We zoom in on the two most extensive assessments. These LT specifications do not only involve experiments with tools and techniques but also an extensive consultation round with stakeholders from public organisations, academia and industry, in order to gather insights into scenarios and best practices. The LT specifications concern (1) the field of automated anonymisation, which is motivated by the need of public and other organisations to be able to store and share data, and (2) the field of multilingual fake news processing, which is motivated by the increasingly pressing problem of disinformation and the limited language coverage of systems for automatically detecting misleading articles. For each specification, we set up a corresponding proof-of-concept software to demonstrate the opportunities and challenges involved in the field. Keywords: anonymisation, fake news, machine translation 1. Introduction The development of language technologies (LTs) re- quires substantial amounts of language data. European, national and regional public services in all EU Member States continuously deal with a huge amount of mul- tilingual textual information in original and translated form. With the European Language Resource Coor- dination (ELRC)1 action, the European Commission (EC) has taken a decisive step towards minimising lan- guage barriers across the EU, through various types of activities: data collection, setup of a helpdesk for data providers, awareness raising, and assessing tools and techniques in specific LT areas. Within ELRC, the as- sessment of tools and techniques in these areas involves documenting them in a hands-on manner and perform- ing corresponding tests and experiments, in order to in- form EC staff and EU Member State representatives about the feasibility and impact of such technologies. The overall goal is to create proof-of-concept environ- ments integrating relevant tools and services in order to facilitate their uptake by users from the public sector. This paper aims to illustrate the development of two major LT assessments in ELRC, i.e. the Automated Anonymisation specification and the Multilingual Fake News Processing specification. The paper is structured as follows. In Section 2, we provide details on the ELRC action, the activities performed in it, and the methodology employed as part of the development of the LT specifications. In Sections 3 and 4, we zoom in on the aforementioned two specifications: (partially) 1http://www.lr-coordination.eu automated anonymisation of monolingual or bilingual information and fake news processing with a multilin- gual focus. Finally, we provide conclusions in Sec- tion 5. 2. Overview of ELRC The ELRC action was set up through …

Content truncated to 500 words