Generative Lexicon Theory and Linguistic Linked Open Data

In this paper we look at how Generative Lexicon theory can assist in providing a more thorough definition of word senses as links between items in a RDF-based lexicon and concepts in an ontology. We focus on the definition of lexical sense in lemon and show its limitations before defining a new model based on lemon and which we term lemonGL. This new model is an initial attempt at providing a way of structuring lexico-ontological resources as linked data in such a way as to allow a rich representation of word meaning (following the GL theory) while at the same time (attempting to) re-main faithful to the separation between the lexicon and the ontology as recommended by the lemon model.