Exploiting verb similarity for event modelling

Abstract

The present thesis aims at exploring the potential of verb similarity, and more specifically of verb classifications, when it comes to capturing and modelling basic information related to events expressed in Spanish. Our research is organised around two studies that examine the ability of verb similarity to model event participant information. We first perform a study of verb similarity with respect to argument structure, looking at its relevant characteristics through the lens of three different perspectives that deal with it (linguistic theory, corpus linguistics and psycholinguistics). This analysis motivates our choice of features and configurations to be explored in the creation of an automatic classification of verb senses using a clustering algorithm. This automatic classification aims at capturing the argument structure of the verbs and reflecting it in the classes in a way that allows for adequately modelling the participants in the events expressed by those verbs. The evaluations carried out for this verb classification confirm the ability of automatic classifications to capture and infer relevant information related to participants in events.