Computational verb lexicons are key to supporting Natural Language Processing (NLP) systems goals for semantic interpretation. Verbs express the semantics of an event being described and relational information among participants, and they project syntactic structures encoding that information, all while displaying a rich range of semantic and syntactic behavior. Verb classifications organize verbs into groups sharing core semantic and syntactic properties. This chapter reviews one such lexical resource for English verbs, VerbNet, which identifies semantic roles and syntactic patterns characteristic of verbs in each class and makes explicit connections between the syntactic patterns and underlying semantic relations that can be inferred for all members of the class. The chapter includes a review of recent updates to the set of thematic roles and a description of the thematic role hierarchy. Ongoing efforts to provide a more empirical grounding of syntactic variations are described, as well as other areas for improvement.