A comparison of the utility of data mining algorithms in an open distance learning context

Abstract

The use of data mining within the higher education context has, increasingly, been gaining traction.A parallel examination of the accuracy, robustness and utility of the algorithms applied to data mining is argued as a necessary step toward entrenching the use of EDM. This article provides a comparative analysis of various classification algorithms within an Open Distance Learning institution in South Africa. The study compares the performance of the ZeroR, OneR, Naïve Bayes, IBk, Simple Logistic Regression and the J48 in classifying students within a cohort over an eightyear time span. The initial results appear to show that, given the data quality and structure of the institution under study, the J48 most consistently performed with the highest levels of accuracy.

(Re) Learning to Teach in Contexts of Crises

“…In South Africa, while both of these concepts (Learning Analytics and Academic Analytics) are still in development, contextual priorities such as focusing on students' access and success, have led to a stronger focus on Learning Analytics (e.g., Lemmens & Henn 2016). Of the sparse literature on Learning and Academic Analytics in South Africa, the majority stems from the University of South Africa (UNISA), which, because of its size and distance education orientation, has had to find ways of managing Big Data (Fynn & Adamiak 2018;Prinsloo, Archer, Barnes, Chetty & van Zyl 2015), as well as considering the ethical implications of data analytics at such scale (Fynn 2016;Willis, Slade & Prinsloo 2016). Beyond UNISA, other recent publications on Learning or Academic Analytics focus on developing models for predicting students' academic performance or to guide enrolment planning (Bleazard & Lourens 2015;van der Merwe, Kruger & du Toit 2018), and developing models or frameworks to guide university teachers to support student success (Janse van Vuuren 2020; Leppan, van Niekerk & Botha 2018).…”