[Corpora-List] Monnet Challenge - Publishing Language Resources as Linked Data

Paul Buitelaar paul.buitelaar at deri.org
Mon Aug 13 18:29:16 CEST 2012

The Monnet Project (http://www.monnet-project.eu/) is offering the following bounties for the conversion of existing linguistic resources into linked data, in particular focussing on the lemon model (http://www.lemon-model.net ). The results of which will be presented at the Workshop on Multilingual Linked Data for Open Enterprise (http://sabre2012.infai.org/mlode) on the 25th September in Leipzig, Germany.

**Bounties are 600, 400, 200, 100, 50 Euros** .

We are targeting the following resources:

- Mapping of a resource containing semantic roles to RDF and/or lemon. Resources such as FrameNet or VerbNet are here considered to be of particular importance - Conversion of existing resources in standard formats to RDF and/or lemon. In particular, we would be keen to see the conversion of a resource in LMF (http://www.lexicalmarkupframework.org/) into lemon. - Creation of lemon lexica for existing Semantic Web ontologies - Extension of DBpedia's Wiktionary (http://wiki.dbpedia.org/Wiktionary) mapping, in particular in the following ways: * Extension of the wrapper templates to new language versions of Wiktionary. * Better extraction of morphological information from the Wiktionary pages, in particular extraction of irregular noun plurals, verb forms and adjectival forms.

Prizes will be awarded based on the following criteria:

Core criteria: * Number of triples (relative to other submissions). Emphasis is of course on number of triples containing a URI from lemon. * Expressiveness and quality of lemon used (How many properties and classes of lemon are you using? Are you using them correctly?) * Impact (Is the data set you converted important and central to our cause? We also rate data sets for less-spoken languages higher, because of the rarity effect.)

Additional criteria: * Note that you can convert and submit more than one data set. You will be rated for the combined data you converted (so each person can only make one submission). * You will be given extra points if you publish converted data early and other people build upon your work (e.g. fix errors). * All submissions will be considered for inclusion in the data post-proceedings.

Submission will end 10 days before the workshop, so the deadline is September 13th, 2012, please submit approximately 2-4 pages documenting your resource to jmccrae -AT- cit-ec.uni-bielefeld.de. In addition your data should be registered at http://thedatahub.org/.

Detailed information on how to submit can be found on the Monnet Challenge page: http://sabre2012.infai.org/mlode/monnet-challenge


John McCrae AG Semantic Computing, Cognitive Interaction Technology - Centre of Excellence Bielefeld University

-------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/html Size: 4195 bytes Desc: not available URL: <https://mailman.uib.no/public/corpora/attachments/20120813/755dbbe8/attachment.txt>

More information about the Corpora mailing list