The Open Linguistics Working Group, in conjunction with the W3C OntoLex Community Group, is organizing the Third Workshop on Linked Data in Linguistics (LDL-2014), in co-location with Language Resource and Evaluation Conference in Reykjavik, Iceland on the 27th of May 2014.  LDL-2014 is also supported by two recently started EU Projects: LIDER (Linked Data as an enabler of cross-media and multilingual content analytics for enterprises across Europe), which aims to provide an ecosystem for the establishment of linguistic linked open data, as well as media resources metadata, for a free and open exploitation of such resources in multilingual, cross-media content analytics across Europe. Secondly,  QTLeap (Quality Translation with Deep Language Engineering Approaches), which explores novel ways for attaining machine translation of higher quality that are opened by a new generation of increasingly sophisticated semantic datasets (including Linked Open Data) and by recent advances in deep language processing.

The goal of the workshop is twofold. First, we will assemble researchers from various fields of linguistics, natural language processing, knowledge management and information technology to present and discuss principles, case studies, and best practices for representing, publishing and linking mono- and multilingual linguistic and knowledge data collections, including corpora, grammars, dictionaries, wordnets, translation memories, domain specific ontologies etc. In particular, we will discuss the application of the Linked Open Data paradigm to linguistic data as it might provide an important step towards making linguistic data: i) easily and uniformly queryable, ii) interoperable and iii) sharable over the Web using open standards such as the HTTP protocol and the RDF data model [1].

Secondly, we will provide researchers on natural language processing and semantic web technologies a platform to present case studies and best practices on the exploitation of linguistic resources exposed on the Web for Natural Language Processing applications, or other content-centered applications such as content analytics, knowledge extraction, etc. The availability of massive linked open knowledge resources raises the question how such data can be suitably employed to facilitate different NLP tasks and research questions. This workshop will also present several contributions to the Linguistic Linked Open Data (LLOD) cloud [2], in particular, contributions that demonstrate an added value resulting from the combination of linked datasets and ontologies as a source for semantic information with linguistic resources published according to as linked data principles. Another important question that will be addressed in the workshop is how Natural Language Processing techniques can be employed to further facilitate the growth and enrichment of linguistic resources on the Web.