Comparison in the application of classification methods to determine the mode of transportation of students to access the Rodrigo Facio campus of the Universidad Costa Rica in Montes de Oca, San Jose, Costa Rica

 

Enregistré dans:
Détails bibliographiques
Auteurs: Hernández Vega, Henry, Sanabria Barboza, Diana
Format: artículo original
Statut:Versión publicada
Date de publication:2022
Description:This work presents the results of an exploratory process where different classification methods were applied to determine the mode of transportation for students to access the Rodrigo Facio campus of the University of Costa Rica. Among the analyzed models are binomial logistic regression, linear discriminant analysis, decision trees, K-closest neighbors, vector support machines and neural networks. A validation was carried out with the K-folds method and a precision higher than 83% was obtained for all the models analyzed. Similarly, the stacking assembly model was applied for the decision tree techniques, K-nearest neighbors, vector support machines, random forests, Bootstrap aggregation, binomial logistic regression, and the potentiation method, obtaining precision values higher than 86% in all cases. The random forest method gives the highest precision.
Pays:Portal de Revistas UCR
Institution:Universidad de Costa Rica
Repositorio:Portal de Revistas UCR
Langue:Español
OAI Identifier:oai:portal.ucr.ac.cr:article/48240
Accès en ligne:https://revistas.ucr.ac.cr/index.php/vial/article/view/48240
Mots-clés:mode of transport
multivariate analysis
ensemble methods
classification methods
modo de transporte
análisis multivariado
métodos de ensamble
métodos de clasificación