Uma abordagem para análise automática de repositórios de objetos de aprendizagem através de técnicas de aprendizagem de máquina para apoio a recomendação personalizada de conteúdo em sistemas adaptativos e inteligentes para educação

• Autor(a) principal: Mendes, Miller Miranda
• Data de Publicação: 2018
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Repositório Institucional da UFU
• Texto Completo: https://repositorio.ufu.br/handle/123456789/21534; http://dx.doi.org/10.14393/ufu.di.2018.233
• topic: Objetos de aprendizagem; Aprendizado de máquina; Learning objects; Estilos de aprendizagem; Learning styles; Felder e Silverman; Adaptive educational systems; Clustering; Learning environment; Lom field; Sistemas educacionais adaptativos,; Ambiente de aprendizagem; Atributos LOM; Computação; Ensino auxiliado por computador; Aprendizado do computador; Agrupamento; CNPQ::CIENCIAS EXATAS E DA TERRA

The amount of digital educational content that is continually created makes the repositories of educational materials increasingly large and the search for specific and targeted content is increasing. Thus, a study was carried out based on learning objects generated in a repository. The IEEE LOM standard was used to generate groups of learning objects that could be used for a given student profile. It is a necessity that these objects be organized so that their recommendation is more efficient. Thus, this paper presents a proposal that uses clustering techniques to group educational content into repositories based on student learning styles. Initially a comparative analysis was performed between grouping algorithms with dummy data in which good results were obtained. Based on these results, a case study using real data from students in an educational system was discussed and implemented.