Review of models for predicting universities student dropout

Abstract

This article aims to show the various efforts made on the research carried out on university dropouts in the world, which merits a comprehensive review. The objective is to be able to identify the variables and the most used machine learning algorithms as well as the proposed models of each researcher. A brief description of the most used learning machines will be made, then a research review will be carried out to later show the most current models proposed. Finally, it is concluded that the most used variables are: University Academic Performance, School Academic Performance, Age, Sex, Dropout and the most used data mining techniques are neural networks and decision trees.