You are here


The Multilingual Central Repository (MCR 3.0) (release 2016)  is a result of the 5th Framework MEANING project (IST-2001-34460) and Spanish government KNOW (TIN2006-15049-C03), KNOW2 (TIN2009-14715-C04-01) projects, SKaTer (TIN2012-38584-C06) projects and the ongoing TUNER (TIN2015-65308-C5). The MCR integrates in the same EuroWordNet framework wordnets from six different languages: English, Spanish, Catalan, Basque,  Galician and Portuguese. The Inter-Lingual-Index (ILI) allows the connection from words in one language to equivalent translations in any of the other languages thanks to the automatically generated mappings among WordNet versions. The current ILI version corresponds to WordNet 3.0. Furthermore, the MCR is enriched with the semantically tagged glosses. The MCR also integrates WordNet Domains, new versions of the Base Concepts and the Top Ontology, and the AdimenSUMO ontology. In that way, the MCR constitutes a natural multilingual large-scale semantic resource for a number of semantic processes that need large amount of multilingual knowledge to be effective tools.

Adimen-SUMO is a translation of SUMO (Suggested Upper Merged Ontology) into an operational first-order ontology. AdimenSUMO can be used appropriately by FO theorem provers (like E-Prover or Vampire) for formal reasoning.

Basic Level Concepts are those WordNet concepts that are frequent and salient; they are neither overly general nor too specific.

Predicate Matrix a new lexical resource resulting from the integration of multiple sources of predicate information including FrameNet, VerbNet, PropBank and WordNet.

QWN-PPV is a method to automatically generate polarity lexicons. It only requires a Lexical Knowledge Base (LKB) such as Wordnet and a list of positive and negative seeds (either words or synsets). The algorithm propagates the initial polarities through the LKB projected over a graph by means of the UKB Personalized PageRank algorithm.

Consistent Annotation of WordNet using Top Ontology is a hierarchy of semantic features designed in the EuroWordNet Project that have been consistently associated to nominal WordNet synsets.

KnowNet is an extensible, large and accurate knowledge base, which has been derived by semantically disambiguating small portions of the Topic Signatures acquired from the web.

deepKnowNet is an extensible, large and accurate knowledge base, which has been derived by semantically disambiguating using page-rank the first 100 words of all the Topic Signatures acquired from the web.

WordNet to FrameNet Mapping is an ongoing work aiming to integrate FrameNet and WordNet.

eXtended WordNet Domains is an ongoing work aiming to automatically improve the domain labels from WordNet Domains.

ImpAr is an algorithm for the annotation of implicit semantic roles.

CrossTimeLines is a crosslingual dataset package to evaluate multilingual and crosslingual timelines (systems, datasets and scripts).

WSD of WordNet Glosses is a small dataset for testing WSD methods on the WordNet definitions.

A2T Domains (A2TD) is a lexical resource generated as part of the Ask2Transformers work.