Comparative Study of Algorithms to Predict the Desertion in the Students at the ITSM-Mexico

In this paper, a comparative study of predicting student dropout risks at the ITSM-Mexico (Instituto Tecnologico Superior de Misantla-Mexico) using SQL server is presented. This system uses the personal and academic information from the students at the ITSM. The comparative study is using four algorithms: logistic regression, clustering, decision trees and neuronal network, these take account the information in the database of the control school system of the institute. The results show that the logistic regression algorithm has a good agreement with experimental results. © 2003-2012 IEEE.

Algorithms; clustering; decision trees; logistic regression; neuronal network Algorithms; Decision trees; Neural networks; Neurons; Regression analysis; Signal encoding; Students; Trees (mathematics); clustering; Comparative studies; Logistic regression algorithms; Logistic regressions; Neuronal networks; School systems; SQL servers; System use; Clustering algorithms

VIVO