Educational data mining in a discipline offered in the distance learning modality

• Autor(a) principal: Fonseca, Enir da Silva
• Data de Publicação: 2020
• Outros Autores: Araújo Jr, Carlos Fernando de, Barbosa, Frederico Kauffmann, Amaral, Luiz Henrique
• Tipo de documento: Artigo
• Idioma: por
• Título da fonte: Research, Society and Development
• Texto Completo: https://rsdjournal.org/index.php/rsd/article/view/7428
• topic: Discovery of knowledge; Data mining; Decision-making process.; Descubrimiento del conocimiento; Procesamiento de datos; Proceso de toma de decisiones.; Descoberta do conhecimento; Mineração de dados; Processo decisório.

Data mining seeks to identify relationships between data, to create information that can generate new knowledge for the development of science, as well as for decision-making by subsidizing new actions that transform the current reality. And the use of educational data mining techniques, which focuses on the development of methods to explore these data sets. In this study, we tried to analyze the accesses and student grades from a virtual learning environment, measuring the evolution according to the number of accesses. Three classes of the mathematical financial discipline offered in the second semester of 2019 were considered for the research, with the data extracted from all the user activities in the content areas, carrying out an analysis according to the CRISP-DM process with implementation in the software. R and RStudio. The results indicate that the longest period of access to the study environment is between Monday and Wednesday, and it has been shown that with a minimum of 55 times of interaction in the environment, students tend to obtain the minimum grade for approval, while that access over 100 times indicate notes close to the maximum value.