Application of AI to predict academic performance and prevent dropout in higher education (RSL) (#1426)

Authors

SALAZAR ALBERTO, RENZO

GUZMAN AQUIJE, ELVIS HENRY

LOZADA FLORES, ROSE MARY

Abstract

This paper presents a systematic literature review on the application of artificial intelligence to predict student dropout in higher education. The author highlights the importance of addressing this problem, since student abandonment not only affects individuals, but also institutional resources and society in general. To address this problem, the application of artificial intelligence is proposed as a tool to anticipate and enhance the academic performance of students. Through the identification of risk factors related to dropout and the provision of personalized intervention strategies, we seek to improve student retention and academic success. The systematic literature review was carried out using the PRISMA methodology and the PICO methodology to define the inclusion and exclusion criteria of the documents. Scopus databases were comprehensively searched and a total of 110 records were identified. After applying the inclusion and  exclusion criteria, 18 academic articles were selected for the systematic review. The results of the systematic review indicate that the application of artificial intelligence can be effective in predicting student dropout and improving student retention. Different machine learning and deep learning models were found that have been used to identify students at risk and offer personalized recommendations. Furthermore, the importance of collecting and analyzing historical data to improve the accuracy of prediction models was highlighted.