Toward an Automatic Extraction of Collocations in Verb Definitions from a Spanish Explanatory Dictionary

Abstract

In this paper we present a method for identifying collocations in an automatic way in verb definitions extracted from the explanatory dictionary of the Royal Spanish Academy, in order to test that collocations can be identified by applying simple heuristics considering only semantic criteria in well-structured textual contexts, as lexicographic definitions are presented. The method identifies candidates for collocations located at the beginning of the definitions that have a special feature: the base of the candidate collocation belongs to the lexical family of the defined verb (1,347 cases). The evaluation of the obtained word combinations was performed both manually and automatically following various statistical and syntactic-semantic criteria. The results of our experiment show that 61% of the extracted verb combinations are collocations, obtaining a recall of 36%.