Hybrid Students’ Academic Performance and Dropout Prediction Models Using Recursive Feature Elimination Technique

Abstract

AbstractPredicting students’ academic performance (SAP) and dropout are crucial aspects of any academic institution management. It enhances the close monitoring of students’ academic progress and helps in averting possible negative consequences of poor students’ performances. Several models had been developed to predict SAP and dropout; however, the sole reliance on predictive accuracy for the decision on classifier’s performance in this domain and the unquenchable search for efficient models left room for the development of novel and more efficient prediction models. This paper presents novel hybrid prediction models that can be used to predict students’ academic performance and dropout using the Recursive Feature Elimination (RFE) technique. The models are hybrid of Support Vector Machine (SVM) and Naïve Bayes (NB) algorithms. The evaluation of the models in comparison with the existing models showed a better and well comparable performance.KeywordsStudents’ Academic PerformanceSupport Vector MachineNaïve BayesRecursive Feature EliminationDropout