Predicting High School Academic Performance in Bosnia and Herzegovina Using Machine Learning Algorithms
Year: 2026
Type: article
Study design: Observational Study
Abstract: This study employs machine learning techniques to predict the academic success of high school students in Canton Sarajevo, aiming to identify key factors influencing performance and assess the accuracy of various predictive models. The results indicate that the Random Forest algorithm demonstrated significant improvements in precision compared to previous studies, highlighting its potential for use in educational decision-making. Key predictors of student performance included prior academic success, subject-specific grades, parental education, and engagement in academic activities such as homework and class participation. The study’s practical implications suggest that educators and policymakers can use machine learning models to develop targeted interventions, improve student guidance, and implement early warning systems for at-risk students.
DOI: 10.21554/hrr.042605
Authors: Edisa Zilic, Mirdin Zilic
Institution: International Burch University
Language: English
Subfield: Computer Science Applications
Field: Computer Science
Domain: Physical Sciences
SDG: Quality education
Open Access status: diamond
OpenAlex ID: W7147198962
FWCI:
Cites:
Cited by:
Related to: