Intelligent Recommendation System for E-Learning using Membership Optimized Fuzzy Logic Classifier

Technology-enhanced learning provides various communication and information technologies for learning and teaching. The teachers also feel comfortable with the Open Educational Resources repositories that learn the existing learning materials when a new course is developed. Yet, this exists as a non-trivial problem because the learning environments should be flexible for the students to learn on the basis of their situations and characteristics. Hence, the main aim of this paper is to provide personalized dynamic and continuous recommendations for online learning systems. This paper plans to implement the novel recommendation system for online learning using intelligent techniques. The main steps of the proposed model are (a) Data collection, (b) Feature extraction, and (c) classification. Initially, the data are collected locally from the Ekhool learning application. Then the feature extraction techniques, such as t-Distributed Stochastic Neighbour Embedding (t-SNE) and Principle Component Analysis (PCA) are used for selecting the most relevant features. Further, the classifier termed as Fuzzy Logic Classifier is adopted as the recommendation system, where the improvement is made in the membership limits by optimizing it with the Rider Optimization Algorithm (ROA). The superiority of the proposed method is proved by the performance analysis in terms of various performance measures.