Date of Publication :2nd September 2024

Abstract:During the COVID-19 pandemic, learners sought a flexible way of learning online in the comfort of their homes, and e-learning systems became very popular. Lecturers can provide learners with course content and training systems through e-learning. Lecturers use the main features such as grading systems, predicting learner performance, uploading course contents, and creating assessments. Machine learning and AI can transform e-learning systems for effective use by lecturers and learners in HEIs. The learners’ learning styles, learning interests, and learner performance can be monitored through these learning systems. A qualitative content analysis will explore which machine-learning algorithms and techniques were used in e-learning systems and their purpose. Machine learning algorithms such as Reinforcement learning can assist with providing personalized learning material according to preferences, while other algorithms used are multilayer perceptron (MP), random forest (RF), K-nearest neighbor (KNN), support vector machine (SVM) and naïve Bayes (NB). Some classification algorithms can predict learners' learning styles in the Higher Educational Institutions (HEIs). K-Nearest Neighbor (KNN) and SVM have been used to predict student performance which is one of the features of a learning management system. Some studies indicated SVM has the best prediction results. One of the issues with using e-learning systems is that lecturers and learners should have the skills to use the technology.

Reference :