Abordagem utilizando realidades aumentada e virtual para suportar cursos baseados em metodologias ativas de aprendizagem (AURAV-SCBMAA)

Abstract

Over the last few years, the evolution of Information and Communication Technologies (ICT) has impacted the interaction between people. In Education, significant changes have occurred in student behavior, and traditional teaching-learning processes have not kept pace with these changes. As a consequence, new learning methodologies have emerged, especially Active Learning Methodologies (ALM), where the student is at the center of the teaching-learning process and no longer the teacher, who becomes the tutor. Courses Based on Active Learning Methodologies (CBALM) lack broad computational support, in particular a Learning Management System (LMS) that meets the specifics of the activities of these courses. The Medicine Programme of the Federal University of São Carlos (UFSCar) employs MAA and, since its inception in 2006, the Ubiquitous Computing Group of the Department of Computing (DC) of UFSCar has been developing projects in partnership with the Department of Medicine (DMed) of UFSCar to provide appropriate computational support for this course. These projects leaded to the development of the product Educational and Academic Management Software for Courses Based on Active Learning Methodologies (EAMS-CBALM), which is the result of a partnership between DC/UFSCar, DMed/UFSCar, Teaching and Research Institute of the Sírio-Libanês Hospital, and a software house. According to the 2019 New Media Consortium (NMC) report, Augmented Reality (AR) and Virtual Reality (VR) are considered emerging technologies to be adopted in Education from 2022 onwards. The main objective of this doctoral project was to investigate the use of AR and VR to support CBALM, in particular at the UFSCar Medicine Programme. For that reason, it was proposed an approach and a software architecture, which uses AR and VR, to guide in the development of computational support for CBALM. This architecture must be coupled to the educational institution's LMS, which in the case of the UFSCar Medicine Programme is the EAMS-CBALM. In order to evaluate the proposed approach and architecture, a prototype was developed for the medical care simulation, to be employed in the Simulation Station and Self-Directed Learning activities of the UFSCar Medicine Programme. For this prototype, were implemented a Virtual Patient to dialogue with the student, and a Virtual Assistant (VA), whose algorithm uses Natural Language Processing to intermediate this dialogue. The VA algorithm obtained and accuracy of 86.21%, slightly lower than 89.66% of the service QnA Maker, but higher than 81.55% of the Machine Learning algorithm.