Modelo do estudante baseado em emoções e perfis de personalidade para recomendação de estratégias pedagógicas personalizadas

• Ano de defesa: 2016
• Autor(a) principal: Melo, Sara Luzia de
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Uberlândia; Brasil; Programa de Pós-graduação em Ciência da Computação
• Palavras-chave em Português: Computação; Personalidade; Inteligência emocional; Emoções; Modelo do estudante; Computação afetiva; Student model; Affective computing; Emotional inteligence; Personality; Emtions; CNPQ::CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO
• Link de acesso: https://repositorio.ufu.br/handle/123456789/18280; http://doi.org/10.14393/ufu.di.2016.517

Emotions affect directly in the learning process. Students who are in a harmful emotion to learning can not assimilate in the best possible way, the content that is offered. Thus, this research proposes a student model based on emotions and personality profiles for the purpose of developing the student affective profile in a Virtual Learning Environment. For a computational validation modeling the classification of Personality profiles and Prediction of Pedagogical Strategies were developed. The classification of personality profiles was performed by an Artificial Neural Network. Furthermore, an experimental analysis to verify the most appropriate training basis from three sets of training was carried out, as well as simulations with different architectures of Neural Network. As a result, there were an accuracy of 97.28 % for general recognition of the nine personality profiles. The definition of pedagogical strategies was conducted by the association of basics emotions with the advice to the proposed educator in Personality Profiles theory. Was used the Decision Tree technique to induce classification rules in order to determine the most appropriate teaching strategies every Personality Profile. Thus, if a harmful emotion to learning is detected, the pedagogical module executes instructions through the information contained in the custom Student Model providing the stimulus student and/or motivation according to their individual characteristics and thus bring it emotion that promotes the learning.