Assembly of classifiers to determine the academic profile of students

The assembly methods, or combination of models, arise with the purpose of improving the accuracy of predictions. An assembly contains a number of apprentices (base models) which, when of the same type are called homogeneous and if of different, heterogeneous. The characteristic is that these apprent...

Descripción completa

Detalles Bibliográficos
Autores: Silva, Jesus, Rojas Plasencia, Karina Milagros, Senior Naveda, Alexa, Barrios, Rosio, Vargas Mercado, Carlos, Medina, Claudia
Tipo de recurso: artículo
Estado:Versión aceptada para publicación
Fecha de publicación:2020
País:Colombia
Institución:Corporación Universidad de la Costa
Repositorio:Repositorio REDICUC
Idioma:inglés
OAI Identifier:oai:repositorio.cuc.edu.co:11323/7790
Acceso en línea:https://hdl.handle.net/11323/7790
https://doi.org/10.1016/j.procs.2020.03.102
https://repositorio.cuc.edu.co/
Access Level:acceso abierto
Palabra clave:Assembly of classifiers
decision trees
artificial neural network
Descripción
Sumario:The assembly methods, or combination of models, arise with the purpose of improving the accuracy of predictions. An assembly contains a number of apprentices (base models) which, when of the same type are called homogeneous and if of different, heterogeneous. The characteristic is that these apprentices do not perform well. The assembly is generated using another algorithm that combines the apprentices, examples of which are the majority vote, the decision table and the neural networks [1]. This article proposes the use of an assembly of classifiers to determine the academic profile of the student, based on the student’s overall average and data related to educational factors.