Uma estratégia baseada em perfis para suporte à formação de grupos de aprendizagem em EAD

• Ano de defesa: 2016
• Autor(a) principal: Cunha , Felipe Oliveira Miranda
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal da Paraíba; Brasil; Informática; Programa de Pós-Graduação em Informática; UFPB
• Palavras-chave em Português: Clusterização; Colaboração; Educação a distância; Formação de grupos.; Clustering; Collaboration; Distance Education; Formation of groups; CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO
• Link de acesso: https://repositorio.ufpb.br/jspui/handle/tede/9257

In literature, much has been discussed about strategies that strengthen the collaborative learning process in Distance Education. Researches have investigated approaches in formation of groups in distance courses, which has allowed the scientific-academic community to try different ways to assign students to groups. However, literature has devoted little to investigate the contributions of distance education tutors to formation of groups. Furthermore, to define an approach in formation of groups to assure that most of the preferences are fulfilled is not an easy pedagogically task, because the characteristics and necessities of each class may vary. In front of this context, this paper aims to define an approach based on groups profiles, specified onwards the distance tutors preferences, to support the process of formation of learning groups. A systematic mapping study was carried out in the literature to identify the best practices in approaches to formation of groups. The profile and preferences of tutors of the course degree in Computer Science from Federal University of Paraíba were investigated. Group profiles were specified for further experiment in homogeneous grouping per Polo, heterogeneous grouping and complementary for programming skills from mathematical modeling and implementation of algorithm grouping. Finally, a case study was held with groups of students to evaluate the proposed approach and to analyze their performance in collaborative activity from distance. The results emphasize that the preferences of distance tutors allowed the definition of a coherent approach of formation of groups in this domain context of this present study. The grouping allowed to create homogeneous, heterogeneous and complementary groups onward educational platform Moodle data. It was applied the size restriction strategy on a partition resulting from the K-means algorithm, allowing to form homogeneous groups of uniform sizes. It was verified an increase in performance by 65.6% of the students in group activity, when compared to the performance in individual activity.