Machine Learning models and feature relevance for student grade prediction
Journal of High School Science · 2024
DOI: 10.64336/001c.115815 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of High School Science, 2024. Verfasst von Izhan Ali. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-SA 4.0).
Abstract (Originalsprache)
Student performance evaluation is a major challenge for most educational institutions including colleges and schools. To improve educational experiences and provide a better learning environment for students, it is important to find the factors primarily responsible for student academic performance. Today’s educational institutions operate in a more complex environment with multiple modes of delivery, engagement, and assessments available. In this work, we performed a two-fold analysis; first to measure the relevance of these factors when considering student grades and second, to use Machine Learning models to predict student grades. The results were then compared along two different evaluation metrics to establish a framework for student grade prediction. Two publicly available datasets were used for this analysis: one from the college level and the other from the school level. The college level data found a greater effect of student engagement on overall performance, whereas the school level data suggested that features such as study time and parents’ job were significant. Romantic relationships were found to affect grades at school level. Two categories of Machine Learning models were used; Linear Models and Ensemble Learning Models. The latter were found to perform well for the grade prediction problem when analyzing college level student grades. Linear models like Linear Regression and Support Vector Regression (SVR) performed better for school level student grades. These results can be used by educators to predict and analyze student performance both at college and school level.
Zitieren
APA
Ali, I. (2024). Machine Learning models and feature relevance for student grade prediction. Journal of High School Science. https://doi.org/10.64336/001c.115815
Vancouver
Ali I. Machine Learning models and feature relevance for student grade prediction. Journal of High School Science. 2024. doi:10.64336/001c.115815
BibTeX
@article{Ali2024,
author = {Ali, Izhan},
title = {Machine Learning models and feature relevance for student grade prediction},
journal = {Journal of High School Science},
year = {2024},
doi = {10.64336/001c.115815}
}
EUVIMED