Modelo de predicción de la deserción universitaria mediante analítica de datos: Estrategia para la sustentabilidad

Laura Guerra, Dulce Rivero, Alexander Ortiz, Eleazar Diaz, Santiago Quishpe

Producción científica: Contribución a una revistaArtículorevisión exhaustiva

2 Citas (Scopus)

Resumen

Data analytics enables organizations to make timely and sound decisions to stay or grow in their markets. Educational institutions must retain their students, because without them the academy does not make sense. This research aims to develop a prototype of a predictive model for university dropout from the evaluation of five data analytics algorithms (KNN, decision tree, random forest, SVM and neural networks), considering the independent variables, characteristics grouped as personal-cognitive, academic-organizational and socioeconomic. It is an applied research, with a mixed focus and where the survey technique and the KDD data mining methodology were applied. The resulting model is the one that is based on neural networks and provides an accuracy of 92%, precision of 90% and f1 of 90%, transforming it into a robust model with a very low percentage of error.

Título traducido de la contribuciónPrediction model of university dropout through data analytics: Strategy for sustainability
Idioma originalEspañol
Páginas (desde-hasta)38-47
Número de páginas10
PublicaciónRISTI - Revista Iberica de Sistemas e Tecnologias de Informacao
Volumen2020
N.ºE35
EstadoPublicada - sep. 2020

Nota bibliográfica

Publisher Copyright:
© 2020, Associacao Iberica de Sistemas e Tecnologias de Informacao. All rights reserved.

Palabras clave

  • Data analytics
  • Neural networks
  • Predictive model
  • Student dropout

Citar esto