Machine Learning-Based Personalized Recommendation System for E-Learners

The epidemic has made taking online courses a popular learning tool for students all over the world. The knowledge over the network is always growing, as well as it is getting hard for learners to seek the necessary knowledge or suitable learning articles to meet his/her desires. Practical 'Personalized Recommendation Systems (PRS)' will not just mitigate such trouble of data overwhelm by suggesting suitable educational articles to the learners, in addition, it offers them correct knowledge in the correct manner and at the correct time. This paper identifies potential directions for future research while methodically discussing the primary recommendation mechanisms used in online learning. Then, we proposed an architecture for developing a machine learning-based PRS for online learners that aim to support learners via navigating lecture videos to reach their point of the interesting topic and to search the related articles to boost the learning skills of the learner. In this paper, the performance of machine learning techniques has been examined, to predict the request-based websites, and discovered that the Random forest approach provides the highest prediction accuracy (98.98%). We have even listed some features of the proposed PRS. Academics and professionals may find the observations in this paper helpful in their efforts to comprehend the present state of play and potential future developments of recommender systems in online learning.