Smart Compendious Resource Recommending E-Learning System Using Concept Tagging

Abstract

The shift to digital forms of learning from books has led to the need for efficient e-learning systems. Today the learning has become so advanced that one resource is not adequate to provide sufficient information to the learner. Thus, a smart resource recommendation system is needed to broaden the learner's concept and make solid conceptual foundations. This paper introduces a smart resource recommending system that predicts useful resources for the learners. The prediction is made based on literature and data-driven method, these methods consider the learning style of a learner for making predictions. Detecting the learning style requires a procedure that is operative and precise. This study makes use of literature based method along with SVM to identify students' learning process. The feature of concept tagging is an additional advantage that gives more personalized predictions based on the understanding of the learner. The resource content is divided based on difficulty levels, a person who already has reached the advanced level of any resource is provided with the suggestion of resources having advanced and complex knowledge, which a person with less grab on the concept would not understand. Thus, it helps all the users by providing the knowledge which is required by the user then since any additional information would be waste. We have used three levels of concept tags, 1. Easy Concept, 2. Moderate and 3. Advanced. The proposed system also aids to increase the knowledge of the learner by providing methods of gamification when the learner needs it. Many surveys show that the gaming approach towards learning aids in strengthening of concepts. Thus, the system provides additional recourse recommendations and also aids in the conceptual understanding of the learner.