A computational approach to Latin verbs: new resources and methods
This thesis presents the application of computational methods to the study of Latin verbs. In particular, the author describes the creation of a sub-categorization lexicon extracted automatically from annotated corpora. Furthermore she presents a probabilistic model for the acquisition of selection preferences from annotated corpora and Latin WordNet. Finally, the author describes the results of a diachronic quantitative study on Latin spatial preverbs.