Integrating XGBoost and Neural Networks for Accurate Student Performance Prediction in Higher Education

Bajrami, Buen and Ristevski, Blagoj and Veljanovska, Kostandina (2025) Integrating XGBoost and Neural Networks for Accurate Student Performance Prediction in Higher Education. In: XV International Conference on Applied Internet and Information Technologies (AIIT 2025), 7 November 2025, Bitola, Macedonia.

Text AIIT2025 Proceedings_Buen.pdf - Published Version Download (2MB)

Abstract

This paper is dedicated to predict student performance as a persistent challenge in the academic world. One of the main reasons why institutions deal with these issues is to help students who are at risk of learning and lack successful results through personalized lessons. We have used and analyzed a hybrid approach consisting of XGBoost (eXtreme Gradient Boosting) and neural networks, which provide quite accurate predictions. This combination through XGBoost and deep learning, offers a higher reliability for building intelligent learning management systems systems (LMS), helping institutions make decisions in order to increase quality and positive results. We have also developed a platform that integrates both of these models and facilitates the practical usage of these cases. We have developed the platform through web technologies, with PHP used for the logical part of the platform and MySQL for data storage and structuring student data.

Item Type:	Conference or Workshop Item (Paper)
Subjects:	Scientific Fields (Frascati) > Natural sciences > Computer and information sciences Scientific Fields (Frascati) > Engineering and Technology > Electrical engineering, electronic engineering,information engineering
Divisions:	Faculty of Information and Communication Technologies
Depositing User:	Prof. d-r. Blagoj Ristevski
Date Deposited:	08 Jan 2026 14:28
Last Modified:	08 Jan 2026 14:28
URI:	https://eprints.uklo.edu.mk/id/eprint/11310

Actions (login required)

View Item