Linked Data for Language Technology (LD4LT) 21 March, Athens, Greece co-located with the European Data Forum 2014. Location: Royal Olympic Hotel, Conference Room 1
Linked Data (LD) has proven beneficial in many new and unforeseen ways for Language Technology (LT) and the newly gained interoperability and availability of LT data and services is currently receiving industry adoption. The Workshop will offer several short talks around the topics of Linguistic resources available in the linked data format and their potential impact on research and industry. We will explore together the advantages of promoting free, open and interoperable language resources as Linked Open Data (LOD) to lower the entry cost to innovating in the use of such resources and technologies in research and industry. We will also aim at exploring how Linguistic Linked Licensed Data open new opportunities for market-based linguistic data exchange.
In summary, we aim at bringing together stakeholders involved in the linguistic data value chain in order to explore new use cases and business models for reusing existing and future language (open or licensed) linked data resources and platforms for the joint development and hosting of multi–language data sets and services in different types of applications such as content analytics, machine translation, sentiment analysis, term disambiguation, etc.
With the foundation of the Linked Data for Language Technologies (LD4LT) W3C community group, this event will start the discussion, analyse current trends and offer a crystallization point to coordinate the development of future LD-based LT applications. The LD4LT Group Kick-Off Meeting and Roadmap meeting is supported by the LIDER EU project, the MultilingualWeb community, the NLP2RDF project, the Working Group for Open Data in Linguistics as well as the DBpedia Project.
As input to the discussion and the work of the LD4LT group, we invite you to take a few short minutes to fill in the first LIDER survey. During the kick-off meeting, via the survey and through the LD4LT group, you will have the opportunity to offer your view on how linked data and language technology should benefit each other.
Session 1 Challenges - Chair: Dave Lewis (CNGL at TCD)
- Dave Lewis (LD4LT Group Co-Chair, CNGL at TCD, Ireland) Welcome and LD4LT Introduction
- Asuncion Gomez-Perez (UPM, Spain): Challenges of Linked Data 4 Language Technology
- Hans Uszkoreit (DFKI, Language Technology Lab, Germany): Language Technologies and Linked Data: Opportunities, Challenges and Synergies
- Nicoletta Calzolari (CNR, Italy): Open Language Resources & Meta-Resources: a Treasure and a Challenge for Linked Data
- Stelios Piperidis (Athena, Greece): Opening and Linking Data and Processing Services
Session 2 Impulses - Chair: Felix Sasaki (DFKI / W3C Fellow)
- Phil Archer (W3C, UK)(W3C Data Activity): Demanding Multilingual Schemata
- Felix Sasaki (DFKI, Germany): ITS 2.0 - Foster Multilingual Schema (and data?) creation
- Arno Scharl (WebLyzard and Modul University, Austria): Semantic Systems for Web Intelligence and Decision Support
- Briefing from W3C Community Groups:
- Philipp Cimiano (CITEC, Universit Bielefeld): Ontology-Lexica Community Group (OntoLex)
- Jorge Gracia (UPM, Spain): Best Practices for Multilingual Linked Open Data Community Group (BPMLOD)
- Gabi Vulcu (INSIGHT, Ireland): Linked Data Models for Emotion and Sentiment Analysis Community Group (SENTIMENT)
Session 3 Position Statements and Industry Perspective - Chair: Philipp Cimiano
Presentations by companies and industrial stakeholders about current problems as well as existing solutions
- Position Statements:
- Athena Vakali (Aristotle University, Greece): From Social Web to the Web of Linked Data
- Martin Benjamin (EPFL: Swiss Federal Institute of Technology): kamusi.org
- Tatiana Gornostay (Tilde): Towards collaboration between Terminology Resources and Linked Data
- John McCrae (CITEC, University Bielefeld, Germany): WordNet: A Case for Linked Data
- Roberto Navigli (Sapienza University of Rome, Italy): The case of BabelNet 2.0
- Costas Nadalis (TMServe, Greece): The Need for Clean Language Data and the Rules for Cleaning it
- Giannis Stoitsis (AgroKnow Technologies, Greece) How can I transform an online service to be truly multilingual?
- Phil Ritchie (VistaTEC and Digital Linguistics, Ireland): Provenance, Metaphor modelling and semantic similarity
- Daniel Vila-Suero (UPM, Spain): Licenses: One of the L's of 3LD
- Ilan Kernermann (K Dictionaries, Israel): Developing multi-language dictionary data for NLP use
- Uroš Milošević (Institute Mihajlo Pupin): Rozeta
Session 4 Requirements and Use Case Discussion - Chair: Dave Lewis
- Georgeta Bordea (INSIGHT, Ireland): Snapshot from LD4LT Online Survey
The LD4LT meeting will be organised by the LIDER Project
- Asuncion Gómez-Pérez (UPM)
- David Lewis (TCD)