Predicción de la Deserción Académica en una Universidad Pública Chilena a través de la Clasificación basada en Árboles de Decisión con Parámetros Optimizados

Abstract

The aim of this study is to present a classification based on decision trees (DTBC) with optimized parameters to predict the dropout rate of university students. The study analyses 5288 cases of students belonging to a Chilean public university. For the CBAD technique, the parameters were optimized to improve the prediction using the software RapidMiner. The result of the application of this technique with optimized parameters achieved a precision rate of 87.27%. It is concluded that the use of DTBC technique with parameter optimization results in a better precision compared to other research with similar number of data.