The idea of this is to work on a roundtrip conversion from the TBX standard for representing terminology to RDF and back. The idea would be to build on the existing code at bitbucket: https://bitbucket.org/vroddon/tbx2rdf
Potential industry partner: TILDE (Tatiana)
Source code: https://bitbucket.org/vroddon/tbx2rdf
TBX Standard: http://www.ttt.org/oscarstandards/tbx/
Contact
person: Philipp Cimiano, John McCrae, Victor Rodriguez-Doncel
The experience on the creation of the Apertium RDF dictionaries will be presented. Taking as starting point a bilingual dictionary represented in LMF/XML, a mapping into RDF was made by using tools such as Open Refine. From each bilingual dictionary three components (graphs) were created in RDF: two lexicons and a translation set. The used vocabularies were lemon for representing lexical information and the translation module for representing translations. Once they were published on the Web, some immediate benefits arise such as: automatic enrichment of the monolingual lexicons each time a new dictionary is published (due to the URIs ruse), simple graph-based navigation across the lexical information and, more interestingly, simple querying across (initially) independent dictionaries.
The task could be either to reproduce part of the Apertium generation process, for those willing to learn about lemon and about techniques for representing translations in RDF, or to repeat the process with other input data (bilingual or multilingual lexica) provided by participants.
Contact person: Jorge Gracia
Babelfy is a unified, multilingual, graph-based approach to Entity Linking and Word Sense Disambiguation. Based on a loose identification of candidate meanings, coupled with a densest subgraph heuristic which selects high-coherence semantic interpretations, Babelfy is able to annotate free text with with both concepts and named entities drawn from BabelNet’s sense inventory.
The task consists of converting text annotated by Babelfy into RDF format. In order to accomplish this, participants will start from free text, will annotate it with Babelfy and will eventually make use of the NLP2RDF NIF module. Data can also be displayed using visualization tools such as RelFinder.
Contact person: Tiziano Flati (flati@di.uniroma1.it), Roberto Navigli (navigli@di.uniroma1.it)