E-learning Recommender System based on k-Nearest Neighbor (KNN), Singular Value Decomposition (SVD), and CoClustering Approaches

The advent of the digital age resulted in the development of the e-learning platform as a helpful resource for top-notch e-learning materials. Despite its promise, the entire scope of its potential hasn't been ultimately discovered. E-learning platforms are expected to offer information that meets customers' requirements and interests to draw users and boost profits. These suggestions are created by considering various aspects, including previous browsing habits, purchase history, demographic data, and others. E-learning platforms may improve the quality of the learning process by offering users interesting information tailored to their specific requirements and preferences by utilizing cutting-edge technology. This paper explores Machine Learning (ML) approaches used in e-learning content recommender platforms. Three ML approaches are chosen and applied to predict user interest and need Singular Value Decomposition (SVD), k-nearest Neighbor Baseline (KNNBaseline), and CoClustering. These ML approaches boost user experience by analyzing data trends and patterns of usage to deliver insights into the best customized and appropriate educational materials for every user. The previous user ratings trend is employed to derive the item ratings. Mean absolute error (MAE) and root mean square error (RMSE) are assessed to evaluate the implemented techniques' effectiveness.