Modelo de computação cognitiva para suporte as atividades educacionais em interações com ambientes inteligentes

• Ano de defesa: 2018
• Autor(a) principal: Santos, Fernão Reges dos
• Orientador(a): Notargiacomo, Pollyana Coelho da Silva
• Banca de defesa: Barbeta, Vagner Bernal , Rossi, Rogério , Lopes, Paulo Batista , Silva, Leandro Augusto da
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Presbiteriana Mackenzie
• Programa de Pós-Graduação: Engenharia Elétrica
• Departamento: Faculdade de Computação e Informática (FCI)
• País: Brasil
• Palavras-chave em Português: computação sensível ao contexto; ambientes inteligentes; sistemas multiagentes; sistemas cognitivos; inteligência artificial
• Área do conhecimento CNPq: CNPQ::ENGENHARIAS; CNPQ::CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO
• Link de acesso: http://dspace.mackenzie.br/handle/10899/24289

The evolution of models and systems that implement the concepts of Ubiquitous and Pervasive Computing, particularly in applications that integrates Context-Aware Computing and mobile computing for personal use, presents opportunities and challenges in the educational field. Parallel to this evolution and in a complementary way, the models of computational intelligence are established as important tools for Internet-based services (solutions that combine applications and services on demand) and support decision systems applying personification to access information in large volumes of data (Big Data platforms). From this scenario, this research proposes a computational model that combines different artificial intelligence methods to aid the personalization of the teaching process considering interaction in context sensitive computing environments. This computational model is, mainly, a proposal of computation with focus on the perspective of the student to participate of educational tasks in different places. As a central element to support the process of customization and adaptability of student interaction, this research presents a computational model that combines Artificial Neural Networks, Genetic Algorithms, Multiagent Systems and Context-Aware Computing. During the investigation of the dynamics of the computational model, an integration of the central model, a set of rules of the Multiagent System and an implementation of the proposed Architecture for Context- Aware Computing were elaborated. The cognitive structure was observed while using by a group of users in different locations. The observed results reinforce the dynamics proposed in this research as a viable model for personalization of interactions with intelligent environments.