Uma Abordagem de Arquitetura em Nuvem para Dados Educacionais em um Sistema Tutor Inteligente em contexto de Big Data

• Ano de defesa: 2022
• Autor(a) principal: Sousa, Regis Michel dos Santos
• Tipo de documento: Tese
• Tipo de acesso: Acesso embargado
• Idioma: por
• Instituição de defesa: Universidade Federal de Uberlândia; Brasil; Programa de Pós-graduação em Engenharia Elétrica
• Palavras-chave em Português: Smart Tutor Systems; Big Data; Cloud Architecture; Microservices; Intelligent Tutoring Systems; Information Retrieval; CNPQ::CIENCIAS EXATAS E DA TERRA; Engenharia elétrica; Computação em nuvem; Big data; Arquitetura de rede de computador
• Link de acesso: https://repositorio.ufu.br/handle/123456789/36697; http://doi.org/10.14393/ufu.te.2022.5039

The teaching and learning process in computer education systems has grown sharply in recent years and Intelligent Tutoring Systems (STI) have gained prominence by propo- sing the customization of environments to the student and thus improving the teaching and learning process. As a consequence of the increase in users of teaching platforms, large volumes of data have been generated, which have been research objectives to pro- pose performance improvement in data retrieval, storage and processing in the Big Data context. This work aims to propose a reference architecture approach in Big Data using cloud microservices for STI-LINA. For this, this approach focuses on mapping the pro- cesses of ingestion, storage, capture and enrichment of data and define adequate services for each of them. As a result of this work, it was possible to propose architectures based on microservices using AWS services, for all with models that make up the STI-LINA (Student, Domain and Pedagogical). In addition, it was possible to propose a new mo- dule/component that will provide analytical information to help the teacher follow the student’s trajectory in the teaching and learning process. Finally, a proposal was made to include the metric access method, called Slim-Tree-Lina, to optimize the recovery of learning objects offered to the student.