An Exploratory Study on the Use of Machine Learning to Predict Student Academic Performance

Patrick Kenekayoro
Additional contact information
Patrick Kenekayoro: Mathematics / Computer Science Department, Niger Delta University, Amassoma, Nigeria

International Journal of Knowledge-Based Organizations (IJKBO), 2018, vol. 8, issue 4, 67-79

Abstract: Optimal student performance is integral for successful higher education institutions. The consensus is that big data analytics can be used to identify ways for achieving better student academic performance. This article used support vector machines to predict future student performance in computing and mathematics disciplines based on past scores in computing, mathematics and statistics subjects. Past subjects passed by students were ranked with state of art feature selection techniques in an attempt to identify any connection between good performance in a particular discipline and past subject knowledge. Up to 80% classification accuracy was achieved with support vector machines, demonstrating that this method can be developed to produce recommender or guidance systems for students, however the classification model will still benefit from more training examples. The results from this research reemphasizes the possibility and benefits of using machine learning techniques to improve teaching and learning in higher education institutions.

Date: 2018
References: Add references at CitEc
Citations:

Downloads: (external link)
http://services.igi-global.com/resolvedoi/resolve. ... 018/IJKBO.2018100104 (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:igg:jkbo00:v:8:y:2018:i:4:p:67-79

Access Statistics for this article

International Journal of Knowledge-Based Organizations (IJKBO) is currently edited by John Wang

More articles in International Journal of Knowledge-Based Organizations (IJKBO) from IGI Global
Bibliographic data for series maintained by Journal Editor ().