Uma metodologia de mineração de dados orientada pelo domínio para a descoberta de conhecimento sobre o processo de aprendizagem no ensino básico

• Autor(a) principal: Fonseca, Stella Oggioni da
• Data de Publicação: 2018
• Tipo de documento: Tese
• Idioma: por
• Título da fonte: Biblioteca Digital de Teses e Dissertações da UERJ
• Texto Completo: http://www.bdtd.uerj.br/handle/1/13713
• topic: Data mining; Educational data; Discovery of knowledge; Learning mathematics; Mineração de dados; Dados educacionais; Descoberta de conhecimento; Aprendizagem matemática; Sistemas de recuperação da informação Educação; Avaliação educacional; Aprendizagem - Avaliação; Matematica - Estudo e ensino; Avaliação de potencial de aprendizagem; CNPQ::CIENCIAS EXATAS E DA TERRA::MATEMATICA::MATEMATICA APLICADA

In the present work, a data mining methodology was developed to identify patterns related to learning of Mathematics and to the characteristics of the school environment, including teachers, principals, and schools where the students are enrolled. More specifically, we looked for aspects that could influence the performance obtained in Mathematics by the students of the 9th year of elementary school who reside in the state of Rio de Janeiro. Data from the Prova Brasil were used, an evaluation developed by the Anísio Teixeira National Institute for Education Research and Studies (INEP), which aims to assess, through the conduction of tests and contextual questionnaires, the quality of basic education. As contribution of the present work we can mention the integrated use of different approaches in order to obtain relevant patterns related to the teaching-learning process. These approaches have been supported by the knowledge of educational experts. In this sense, among the tasks related to the discovery of knowledge, the proposal to use a data dimensionality reduction methodology integrated with the use of measures of technical interest together with measures of educational interest. Such integration allowed an analysis under a broader perspective, addressing actions, as well as the feasibility to solve educational issues. The results presented important factors that can positively and negatively affect students' performance. Finally, the developed methodology for the extraction of knowledge related to the learning process of Mathematics can support other research, providing aspects capable of fomenting and enriching the discussion among decision makers.