The Multilingual Linked Open Data for Enterprises (MLODE 2014) workshop will bring together developers, data producers, academia and enterprises from various fields of linguistics, natural language processing, and information technology (Linked Data) to present and discuss principles, case studies, and best practices for representing, publishing and linking linguistic data collections, including corpora, dictionaries, lexical networks, translation memories, thesauri, etc.
As Semantic Web research progresses, interest by practitioners, industry and infrastructure providers operating across language barriers increases. The Linked Open Data (LOD) community is enthusiastic about the new possibilities offered by new, vast multilingual resources. While it is clear that the Semantic Web is not a panacea, it has matured into a technology capable of addressing specific real world problems of globalisation faced by industry and the governmental sector.
For MLODE 2014, we will invite a mixture of industry, researchers and engineers. Our goal is to compare technologies and datasets developed before and in parallel to linked data, RDF and OWL with an open mind to find out what is the right strategy for the right type of problem.
This focus on effectiveness will allow industrial participants to get a good idea and overview of which technologies are ready for exploitation. The workshop will be very interactive and you are expected to help us achieve common goals:
In the past, MLODE 2012 has successfully created a highly-productive interdisciplinary network, where linguistic researchers exchange experience on how to sustainably manage and interlink their huge amounts of language data most efficiently through Semantic Web technologies whereas on the other side the Linked Data research is provided with scientific data to drive development of more precise models and applications. One major output of MLODE 2012 was the improvement of the technical viability of the Linguistic Linked Open Data (LLOD) cloud: http://sabre2012.infai.org/mlode
This year, MLODE will focus on the topics of “Unifying the Dictionary”, “Building LLOD-aware NLP services” and “Generating Linked Data from Language Resources”, which are of great significance for the areas of linguistics, computational linguistics as well as sciences exploiting NLP methods as a basis for their research, i.e. Cultural Heritage, e-Humanities and other text- and data-rich sciences.
Unifying the Dictionary aims to continue and centralize the various efforts by the community to compile a multilingual dictionary integrating heterogeneous sources. Open questions like focusing the research, chances of crowd-sourcing a multilingual core lexicon and actual use cases of companies will be tackled using the combined experience of the assembled research communities, as well as the industry presence associated with SEMANTiCS. Finally, the DBpedia project is willing to contribute to creation, hosting and maintenance of the final resources, guaranteeing persistence, impact and visibility of the outcome.
Building LLOD-aware NLP services will focus on building a free, open and interoperable repositories of Linguistic Linked Open Data resources and services, including comprehensive structured metadata. The final goal is effortless programmatic access to multilingual NLP services, not unlike the Language Grid project, but based on open data.
Generating Linked Data for Language Resources continues the effort of converting, aggregating and publishing Language Resources made during the last MLODE and aims to extend the LLOD-cloud, which has grown significantly, contributing linguistic data to the NLP service infrastructure: http://linguistics.okfn.org/resources/llod/
Business cases and removal of barriers for industry participation in Linked Data for natural language processing and for content internationalisation and localisation
Discussion on best practices on how to channel feedback from industry to open-source and academic communities
Generate a Roadmap for Linked Data & Language Technology in Europe
The following organisations are contributing to MLODE 2014:
- LIDER (http://lider-project.eu)
- Linked Data for Language Technology W3C Community Group (http://www.w3.org/community/ld4lt/)
- Working Group for Open Data in Linguistics (http://linguistics.okfn.org)
- K Dictionaries (http://kdictionaries.com)
- Ontotext (http://ontotext.com)
- Clarin (http://clarin.eu)
- Language Science Press (http://langsci-press.org/)
- Kernerman DICTIONARY News (http://kdictionaries.com/kdn.html)
- DBpedia Association (http://dbpedia.org/Association)
- InfAI (http://infai.org)
- NLP2RDF (http://nlp2rdf.org)
- Sebastian Nordhoff, OWLG and Language Science Press
- Steven Moran, OWLG and Universtät Zürich
- Christian Chiarcos, OWLG and Universität Frankfurt
- Ilan Kernermann, K Dictionaries
- Marin Dimitrov, Ontotext
- Volker Boehlke, Clarin and University of Leipzig
- Gerhard Heyer, Clarin and University of Leipzig
- Philipp Cimiano, LIDER and CIT-EC (University Bielefeld)
- John McCrae, LIDER and CIT-EC (University Bielefeld)
- Matthias Hartung, LIDER and CIT-EC (University Bielefeld)
- Dave Lewis, LD4LT, LIDER and CNGL at Trinity College Dublin
- Jorge Garcia, LIDER, Universidad Politécnica de Madrid (UPM)
- Sebastian Hellmann, LD4LT, NLP2RDF and AKSW
- Martin Brümmer, OWLG, LIDER and AKSW
- Bettina Klimek, OWLG, LIDER and AKSW