Análise dos cursos de licenciatura da Rede Federal utilizando mineração de dados

• Autor(a) principal: Gualhano, Mariana Abreu
• Data de Publicação: 2020
• Outros Autores: de Vasconcelos, Aline Pires Vieira
• Tipo de documento: Artigo
• Idioma: por
• Título da fonte: Educação e Pesquisa
• Texto Completo: https://www.revistas.usp.br/ep/article/view/187183
• topic: Educational evaluation; Federal Schools; PCC; Data mining; Teaching Certification programs; Avaliação; Rede Federal; CPC; Mineração de dados; Licenciatura

Knowledge extraction, also known as the KDD (Knowledge Discovery in Databases) process, is a set of techniques (Selection, Pre-processing, Treatment, Data Mining and Data Interpretation) that aims to analyze and extract potentially useful patterns and information from large databases. The evaluation of the quality of undergraduate courses in Brazil is done through the Preliminary Concept of Courses (PCC), which is a quality index that evaluates these courses. Therefore, this research is part of the context presented here, seeking to use data mining techniques to extract knowledge from the PCC evaluations of the years 2014 and 2017 and to identify the main criteria and results of the evaluation of Teaching Certification programs, making an analysis of data from Federal Schools all over Brazil. In order to do that, the database available on the website of INEP (National Institute for Educational Studies and Research) and the KDD process steps were used, focusing on Data Mining to extract knowledge from the database. The results enabled the identification of the evaluation criteria with the greatest impact on the evaluation of the PCC of Teaching Certification programs, in addition to a comparison between the evaluation of 2014 and 2017. The information extracted from the present work will hopefully be useful and may subsidize educational management, in addition to being used to improve undergraduate courses.