Use of Artificial Intelligence to Predict University Dropout: A Quantitative Research

Abstract

The main aim of the research is to predict, as early as possible, which student will drop out in the Higher Education (HE) context. Artificial Intelligence (AI) is used for replacing repetitive human activities, e.g. in the field of for autonomous driving or for the task of classification pictures. In these areas IA competes with the man with fairly satisfactory results and, in the case of college dropout, it is extremely unlikely that an experienced teacher can “predict” the student’s academic success based on only on data provided by administrative offices. In this study used administrative data of about 6,000 students enrolled in the Department of Education of the University of Roma Tre to train convolutive neural nets (RNC). The trained network provides a probabilistic indicating, for each student, the probability of abandonment. Then, the trained network provides a predictive model that predicts whether the student will dropout. The accuracy of the obtained deep learning models ranged from 67.1% for the first-year students up to 94.3% for the third-year students.