Automatic evaluation of discursive answers short based on three linguistic dimensions / Avaliação automática de respostas discursivas curtas com base em três dimensões linguísticas

Abstract

As the use of virtual environments grows, there is a need for a system of automatic evaluation of discursive answers. This paper proposes a method for automatic evaluation of discursive short answers based on a machine learning architecture. The predictive method is based on the collection of features (140) of similarity between texts in a taxonomy of three linguistic dimensions: lexical, syntactic and semantic. As a result, we obtained quadratic kappa 0.72 human x system (SxH) against 0.94 human x human (HxH)