Desenvolvimento de modelo de inferência de conhecimento baseado nas expressões faciais e nos comportamentos observáveis.

Abstract

Learning, for a long time, had only reason as basement for its development, but studies prove emotion is an essential part in the process of the construction of knowledge. Currently, computational systems have resources to aid emotion recognition in humans, through facial expressions, for example. In this respect, this work wants to analyze the emotions, captured through facial expressions and analyzed by FaceReader software, together with student’s observable behaviors in a virtual learning environment. Using these data was proposed a model for inference the construction of knowledge considering the exercise performance. As theoretical framework this work introduces definitions of emotion, basic emotions models, and techniques to recognize emotions through computers or electronic devices. The emotions, from facial expressions, and observable behavior, from the interaction with a virtual learning environment, were recorded using a screen recorder’s software and with the webcam. These data were treated and related to show which emotions happened during each observable behavior. This work allowed, besides the creation of a model for inference, the analysis of which behaviors and emotions were more important in the process of the construction of knowledge.