Classificação dos estilos de aprendizagem baseado em sistemas inteligentes: um estudo de caso na educação mediada por tecnologia

Abstract

Distance education (DL) courses, increasingly common thanks to the advancement of the internet and the Virtual Learning Environments (VLE) that serve as interaction tools between student and educator, have become a major research niche in worldwide because of its ability to allow certain content to be taught to different people in different places and at different times. Most courses mediated by technologies in online environments present content in static pedagogical sequence, that is, always the same resources applied in the same order for all users, thinking about this, new methodologies were developed to improve the teaching and learning in distance learning. , such as artificial intelligence methods that seek to find relationships between pedagogical progress and educational technological resources available in VLEs. And these new methodologies become more real if applied in the distance learning modality, with the use of a VLE, as it will allow the same course to be customized for each user. An interesting parameter that could guide this personalization is the Student Learning Style, which identifies characteristics of how students learn. In this context, this thesis sought to associate the learning styles theory, which identifies the learning preferences of each student, with the observed behavior of distance education students through their interactions with VLE, using artificial intelligence techniques. This study aimed to analyze the relationship between the learning styles of a set of students and their behaviors registered by the VLE, aiming to answer the following questions: It is possible, automatically, to identify the students’ LS from their interactions with the student. Virtual Learning Environment (VLE)? What techniques can be developed to identify the students’ AE in courses taken in distance learning mediated by technologies, in order to improve a better academic path for student learning? To answer these questions, we use some artificial intelligence algorithms to identify the relationship of students’ AEs with their AVA behaviors. The results show that there is no relationship between Learning Styles and behavior variables. This difficulty can be explained by flaws in the instrument that identifies Learning Styles or in the variables that are used to characterize behavior, showing a lack of correlation between these theories.