Descrição de formalização de verbos de ação-processo para elaboração de parser

• Ano de defesa: 2009
• Autor(a) principal: Rodrigues, Carlos Roberto de Souza
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal do Espírito Santo; BR; Mestrado em Estudos Linguísticos; UFES; Programa de Pós-Graduação em Linguística
• Palavras-chave em Português: Computational linguistics; Parsing; Verb valency; Action-process verbs; Argument structure; Semantic roles; Lingüística computacional; Valência verbal; Verbos de ação-processo; Estrutura argumental; Papéis temáticos; Lingüística - Processamento de dados; Gramática comparada e geral; Semântica; Linguística; 80
• Link de acesso: http://repositorio.ufes.br/handle/10/3706

Chafe (1970) developed a research program that led to six semantic subcategories for the classification of verbs, one of them being action-process verbs. However, literature on this matter provides a very concise theoretical and methodological framework with respect to both the semantic and syntactic properties of the subcategory in question. In order to make available more syntactic-semantic information on action-process verbs, the present research was designed on the basis of a program of identification of verbal valence proposed by Borba (1996) and Welker (2005). Therefore, we investigated four types of verbal valence: logical (Tesnière, 1959; Helbig and Schenkel, 1975), syntactic (Borba, 1996; Ignácio, 2001), and semantic or syntactic-semantic valence (Fillmore, 1968; Travaglia, 1985; Dik, 1989; Dowty, 1989). At the end of this stage of linguistic research, we confirm the heterogeneity of the subcategory of action-process verbs, which we divided into ten subgroups. Furthermore, we specified the argument structures belonging to subgroups, as well as the contents of the syntactic arguments that compose such syntactic configurations. Eventually, the linguistic knowledge obtained during this study allowed us to construct three language resources that provide the basis for the elaboration of computational resources for natural language processing: (i) a lexicon-grammar table, containing the morphosyntactic-semantic properties of verbs and of their syntactic arguments, (ii) a lexical data base, with the morphosyntacticsemantic properties of the verbs in question, and (iii) the argument structures identified in each subgroup.