Feature Evaluation of Emerging E-Learning Systems Using Machine Learning: An Extensive Survey

Abstract

Emerging with the dawn of Artificial Intelligence (AI) and Machine Learning (ML), nowadays, much adaption in e-learning systems would be on diverse learning needs. The present study focuses on the adaptivity levels of students on e-learning platforms through the use of machine learning techniques, mainly in predicting whether the adaptivity level of a student is high, moderate, or low. There are many machine learning algorithms, such as Decision Trees, Logistic Regression, and Random Forest, that have been analyzed for this purpose, but the study proposes Gaussian Naive Bayes (GaussianNB) because it is simple, effective, and computationally efficient in processing the classification problem at hand. GaussianNB is a probabilistic classifier based on Bayes' theorem that provides an appropriate solution in terms of resource and processing constraints for adaptation modeling as it makes an effective handling of continuous data and is comparatively cheaper in terms of computational cost. The performance of GaussianNB against other techniques is evaluated, proving how well it does in terms of delivering precise, speedily computable, and interpretable predictions. The results lead to the conclusion that GaussianNB is a practicable answer to boosting e-learning systems toward student responsiveness and improving their overall drive towards experiential learning.