Using Artificial Neural Networks for Predictive Modeling of Graduates' Professional Choice

Abstract

This paper presents the results of research carried out at the Faculty of Mechanical Engineering (FME) in Belgrade for the purpose of enrollment policy. Artificial neural networks are used in predicting graduates' professional choice, i.e. a number of secondary vocational education and training schools (VET schools) graduate students will enroll on. The assumption of graduates' professional choices was verified on a sample of 119 graduates from two Belgrade VET schools. Factors influencing the professional choice of VET school students are grouped in nine input variables. The results show that neural network algorithms present a powerful tool for predicting graduates' professional choice.