Personalized and Optimized Recommender System for E Learners Based on their Choice and Learning Level

Abstract

Web-based learning is often called online learning or e-learning because it newlineincludes online course content. Many of e-learning systems don t have newlinepersonalisation based on individual needs and their capabilities. Main newlinechallenging aspect of personalized delivery of e-learning is concerned with an newlineadaptive course delivery along with content delivery. Personalized e-learning environment provide recommendations to learning newlinecommunity for supporting and also helping them go through the process of e- newlinelearning, as it plays a crucial role in promotion of smart cities through smart newlinelearning. Personalization of the e-learning systems based on the requirement newlineand knowledge level of learner acts as an important element in a learning newlineprocess. E-learning models with personalized recommendation should adjust newlinethe learning experience with respect to the goal of the target learner.The recommender system given for personalized recommendations is on a set newlineof objects and their utility to a certain domain which begins from the available newlineinformation on objects and users. In this research study, three stage model called Clustering Classification and newlineRecommendation (CCR) is proposed. The research work carried out newlineintroduces a novel model to provide personalized and optimized resources to newlinethe learners based on their capability and level of interests. An innovative newlineframework, Personalized Bee Recommender for e-learning (PBReL) built on Artificial Bee Colony (ABC) optimization was anticipated that utilizes the K- newlineMeans clustering. Experimentations were conducted by employing web-links newlinealong with the elements of the Moodle Learning Management System (LMS). newlineOutcomes illustrated that the projected framework attained sophisticated newlineprecision as well as coverage. A Memetic Swarm Clustering (MSC) technique newlineincorporating ABC and PSO with the K-Means algorithm was introduced to newlinecategorize the choice-based learners and classify them into slow, medium, and newlinefast learners through the use of Deep Belief Network (DBN).