Overcoming language barriers

What is this action about?

Language barriers in the EU make the European market fragmented and decrease its economic potential. The EU institutions aim to overcome language obstacles and increase cross-border e-commerce by building open multilingual tools and features free of charge. This public multilingual knowledge management infrastructure will support e-commerce solutions, such as machine translation, localisation and multilingual search, by creating interoperable multilingual classifications and terminologies that will be easily reusable by small and medium-sized enterprises (SMEs) and public administrations. SMEs are currently at a disadvantage compared to big companies due to the high cost of providing multilingual services.

What are the objectives?

The main objective is to help SMEs and public administrations save resources on creating and maintaining their own multilingual tools.

What are the benefits?

For the EU economy

• Increased cross-border e-commerce
• Easier localisation for e-commerce platforms
• Better terminology quality and harmonisation
• Easier implementation of multilingual search capabilities

For the EU language technology industry

• Cost reduction and faster implementation of cross-border e-commerce services
• Opportunities to use resources saved on building individual multilingual tools for innovation

For the Member States

• Internationalisation of public administration e-services
• Increased ability of public administrations to exchange and use data based on shared classifications and terminologies
• Possibility to extend existing classifications and terminologies through increased interoperability

For EU institutions and other EU bodies

• Increased interoperability and visibility of existing multilingual classifications and terminologies
• Increased interoperability of multilingual Linked open data, i.e. structured interlinked data
• Support of the CEF Automated Translation platform

What has been already achieved?

In 2017, the core data model for multilingual taxonomies/terminologies based on existing standards (Ontolex-Lemon) has been elaborated and reviewed by external experts. 

• PMKI data model
• PMKI data model documentation

In 2018, semantic links between different language resources have been established, in particular two gold-standard alignment datasets between the European Union thesaurus EuroVoc and two other notable resources adopted in legal environments:

• EuroVoc and the thesaurus of the Italian Senate TESEO
• EuroVoc and the IATE European terminological resource
• More details can be found in the Criteria, protocols and results for assessing the mapping methodologies on the identified gold-standard (report).

Several scientific papers has been published:

1. International journals
   • P. Schmitz, F. Sanmartin, E. Francesconi, N. Hajlaoui, B. Batouche, Automatic Alignment of Multilingual Resources in the Linguistic Linked Open Data Cloud, in Journal of Open Access to Law 
   • M. Fiorelli, A. Stellato, T. Lorenzetti, A. Turbati, P. Schmitz, E. Francesconi, N. Hajlaoui, B. Batouche. Towards OntoLex-Lemon editing in VocBench 3, in AIDAinformazioni, Rivista di scienze dell'informazione, ISBN 978-88-548-8992-7, ISSN 1121-0095.  
2. International conference
   • A. Stellato, A. Turbati, M. Fiorelli, T. Lorenzetti, P. Schmitz, E. Francesconi, N. Hajlaoui, B. Batouche, Towards the Assessment of Gold-Standard Alignments between Legal Thesauri, in JURIX 2018 - The 31st international conference on Legal Knowledge and Information Systems. December 12–14, 2018 in Groningen, Netherlands.
   • P. Schmitz, E. Francesconi, N. Hajlaoui, B. Batouche, A. Stellato, Semantic Interoperability of Multilingual Language Resources by Automatic Mapping',  in Proceedings of the 7th International Conference on Electronic Government and the Information Systems Perspective (EGOVIS 2018), Regensburg, Germany 3-6 September 2018. 
   • Schmitz P., Francesconi E., Hajlaoui N., Batouche B., PMKI: an European Commission action for the interoperability, maintainability and sustainability of Language Resources in Proceedings of the 11th edition of the Language Resources and Evaluation Conference, 7-12 May 2018, Miyazaki (Japan).
   • Stellato A., Fiorelli M., Turbati A., Lorenzetti T., Schmitz P., Francesconi E., Hajlaoui N., Batouche B., Dataset Alignment and Lexicalization to Support Multilingual Analysis of Legal Documents in AI Approaches to the Complexity of Legal Systems - Models and Ethi- cal Challenges for Legal Systems, Legal Language and Legal Ontologies, Argumentation and Software Agents, within JURIX 2017 Conference, University of Luxembourg, 13-15 December 2017.
   • Schmitz P., Francesconi E., Hajlaoui N., Batouche B., Towards a Public Multilingual Knowledge Management Infrastructure for the European Digital Single Market, in Proceedings of the LDK 2017 Workshops: 1st Workshop on the OntoLex Model (OntoLex-2017), Shared Task on Translation Inference Across Dictionaries & Challenges for Wordnets, co-located with 1st Conference on Language, Data and Knowledge (LDK 2017), pp. 33-42, CEUR Workshop Proceedings Vol-1899, ISSN 1613-0073, Galway, Ireland, 8 June 2017. 
   • M. Fiorelli, A. Stellato, T. Lorenzetti, A. Turbati, P. Schmitz, E. Francesconi, N. Hajlaoui, B. Batouche, "Editing OntoLex-Lemon in VocBench 3", in Proceedings of the 12th International Conference on Language Resources and Evaluation (LREC 2020), Marséille, France, 11-16 May 2020
   • Best Paper Award: M. Fiorelli, A. Stellato, T. Lorenzetti, P. Schmitz, E. Francesconi, N. Hajlaoui and B. Batouche, ``Metadata-driven Semantic Coordination'', in 13th International Conference on Metadata and Semantics Research, Rome, Italy, 28-31 October 2019.
      

What has been achieved?

• A core data model based on Ontolex-Lemon and a first set of extensions have been defined and validated by external experts (including documentation).
• The technical architecture of the PMKI platform has been defined and validated by external experts.
• A proposal for an adequate government structure has been elaborated.
• The feasibility study for the enhancement of the semantic capabilities of the platform has been completed.
• The implementation of the operational prototype of the PMKI platform is well-advanced. An intermediate version has been presented to the Steering Committee in June 2019 and has been tested by the Publications Office.
• The communication and dissemination activities are proceeding as scheduled and the related key performance indicators targeted for 2016, 2017 and 2018 as well as for the first half of 2019 were achieved.
• The documentation of the PMKI prototype was published. 

What are the next steps?

The next step will be the finalization of the operational prototype of the PMKI dissemination layer based on the defined functional requirements.

The current results of the action and their recognition by potential stakeholders confirm that there is a strong interest in a PMKI platform as a public service. In particular, the combination of multilingual assets from different origins (EU Institutions, international organisations, public administrations and well-recognised public resources on the Web) will create the conditions for a long-term sustainability of the PMKI platform. In consequence, the final system will be installed as a public service in the cloud. It will be managed by the Publications Office and it will be open for contributions of external partners that will also be involved in the governance of the platform.