MODELO PARA DETERMINAR PERFIS DE DESEMPENHO ACADÊMICO NA UNNE COM MINERAÇÃO DE DADOS EDUCACIONAIS

• DOI: 10.22533/at.ed.6132004096

• Palavras-chave: academic performance; data warehouses; data mining; predictive models.

• Keywords: academic performance; data warehouses; data mining; predictive models.

• Abstract:

  Abstract

  In this work it is proposed to evaluate the performance of the students using Data Mining techniques. The proposal does not focus on analyzing the student's profile only through their grades, but also studying academic performance based on other variables. To define the profiles of the students and determine patterns that lead to academic success or failure, we will implement a model that relates the students' qualifications with other variables, such as socioeconomic, demographic, and attitudinal factors, among others; based on which we will classify the different student profiles.

  We describe the model to be implemented with the use of the Data Warehouse to determine the academic performance profiles in the Algebra subjects of the Bachelor's Degree in Information Systems (LSI) of the Faculty of Exact and Natural Sciences and Surveying (FaCENA) of the National University of the Northeast (UNNE) and Mathematics I of the Agricultural Engineering (IA) degree of the Faculty of Agricultural Sciences (FCA) of the UNNE (PI 16F002 accredited by Res. No. 970/16 CS).

  We hope to contribute to finding an answer to the historically observed low academic performance of students, a problem that is the trigger for our research. The predictive models we are looking for will allow us to take actions aimed at avoiding academic failure, detecting students with a risk profile of academic failure early, shortly after the start of the course; This will make it possible to focus on them the efforts of tutoring and special supports.

• Número de páginas: 13