Different lexical resources may pursue different views on lexical meaning. However, all of them deal with lexical items as common basic components, which are described according to criteria that may vary from one resource to another. In this paper, we present a method for measuring the degree of similarity between a valency-based lexical resource and a WordNet. This is motivated by both theoretical and practical reasons. As for the former, we wonder if there are lexical classes that "impose" themselves regardless of the fact that they are explicitly recorded as such in source lexical resources. As for the latter, our work wants to contribute to the research task dealing with merging lexical resources. In order to apply and evaluate our method, we propose a normalized coefficient of overlapping that measures the overlapping rate between a valency lexicon and a WordNet. In particular, in the context of the exploitation of the linguistic resources for ancient languages built over the last decade, we compute and evaluate the overlapping between a selection of homogeneous lexical subsets extracted from two lexical resources for Latin.
|Numero di pagine||19|
|Stato di pubblicazione||Pubblicato - 2016|