Abstract: The number of e-learning platforms and blended learning environments is continuously increasing and has sparked a lot of research around improvements of educational processes. Here, the ability to accurately predict student performance plays a vital role [5]. Specialized scope classifiers are then combined to an ensemble to robustly predict student learning behavior on learning objectives independently of the students’ individual learning setting [7]. The study aims to predict students learning behavior based on students’ interactions with the virtual learning environments [6]. Firstly predictive models are built using traditional classifiers Logistic Regression, Support Vector Machine which are shown good performance. Later implemented popular ensemble methods based on bagging and boosting. The result revealed that the accuracy, precision, recall and F1 score had been considerably improved by ensemble bagging and boosting methods than that of the individual classifiers. The data used in the study is the Open University Learning Analytics dataset (OULAD) set [13] of year 2014.
Keywords: Classification, Machine learning, Virtual learning environment. Ensemble methods, Bagging, Boosting
|
DOI:
10.17148/IARJSET.2022.9735