Arquitetura de um sistema tutor inteligente para recomendação personalizada de objetos de aprendizagem considerando os estados afetivos e o conhecimento do estudante

• Ano de defesa: 2021
• Autor(a) principal: Melo, Sara Luzia de
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Uberlândia; Brasil; Programa de Pós-graduação em Engenharia Elétrica
• Palavras-chave em Português: Sistemas Tutores Inteligentes; Intelligent Tutoring Systems; Computação Afetiva; Affective Computing; Personalidade; Personality; Planejamento Adaptativo do Curso; Adaptive Course Planning; Instrução Adaptativa; Adaptive Instruction; Informática na Educação; Informatics in Education; CNPQ::ENGENHARIAS::ENGENHARIA ELETRICA; Engenharia elétrica; Informática na educação; Inovação pedagógica; Educação afetiva; Ensino, aprendizagem e tecnologias
• Link de acesso: https://repositorio.ufu.br/handle/123456789/33438; http://doi.org/10.14393/ufu.te.2021.559

This work proposes an architecture of an Intelligent Tutoring System based on the Theory of Structured Knowledge Maps and on the Effective Exponential Memorization Method in the Binary Base to customize the learning objects suggested to the student. Thereunto, it is proposed to include the teacher interface in the system architecture. The adaptive sequencing of the course is carried out through the theory of Structured Knowledge Maps, in which the teacher is responsible for specifying the minimum concepts and knowledge needed to understand each item in the syllabus. In the student interface, in order to avoid cognitive overload, the system maps the doubts related to prerequisites of concepts and knowledge, and then presents the learning objects, in different formats, according to the learning needs of the student. Another problem investigated in this work is the adaptability of pedagogical strategies in computational learning environments. When considering the student's personal characteristics in the cognitive process, the student model of this approach, in addition to the knowledge, includes the student's affective profile that contains information about emotions and personality attributes. Thus, the adaptation of instructions and/or motivation takes place considering the different personality profiles, in order to bring the student closer to the emotion that stimulates their learning skills. Therefore, through the information contained in the Student Model and the Domain Model, upon detecting that the student is in an unproductive learning cycle, as well as in an emotion that is detrimental to the learning process, the Pedagogical Module executes instructions personalized to the student's prior knowledge and affective profile.