Construction and implementation of a model to predict the academic performance of university students using the Naïve Bayes algorithm

One of the most widely used applications of educational data mining is predicting academic performance. The aim of this paper is to present the construction, evaluation and implementation of a predictive model of the academic performance of university students by means of the data mining technique k...

Descripción completa

Detalles Bibliográficos
Autores: Rico Páez, Andrés, Gaytán Ramírez, Nora Diana, Sánchez Guzmán, Daniel
Tipo de recurso: artículo
Estado:Versión publicada
Fecha de publicación:2018
País:México
Institución:UNIVERSIDAD DE GUADALAJARA
Repositorio:Diálogos sobre educación. Temas actuales en investigación Educativa
Idioma:español
OAI Identifier:oai:dialogossobreeducacion.cucsh.udg.mx:article/509
Acceso en línea:http://dialogossobreeducacion.cucsh.udg.mx/index.php/DSE/article/view/509
Access Level:acceso abierto
Palabra clave:prediction – academic performance – data mining – predictive model – Naïve Bayes algorithm
predicción
rendimiento académico
minería de datos
modelo predictivo
algoritmo Naïve Bayes
Descripción
Sumario:One of the most widely used applications of educational data mining is predicting academic performance. The aim of this paper is to present the construction, evaluation and implementation of a predictive model of the academic performance of university students by means of the data mining technique known as the Naïve Bayes algorithm. We collected data from 122 students as training for the algorithm and applied the model to predict the academic performance of 71 students. The results show that, in addition to obtaining predictions of academic performance, the predictive model also identifies the factors that influence it the most. This type of study allows teachers to design prevention strategies and identify students who are vulnerable to failure.