الفهرس 
Title : An Enhanced Approach for a Personalized E-Learning Environment using Recommender System
AbstractOnly 14 pages are availabe for public view
 |  | from | 114 |  |  |
| | | from | 114 | | |
Abstract
Substantial improvements in the era of modern technology have led to the development of numerous online learning materials. Sorting these online learning materials to meet the needs and goals of learners is quite difficult due to information overload. However, difficulties pertaining to information overload can be addressed using recommender systems. A recommender system provides an intelligent solution to information overload problem by applying filtering techniques that provide the most useful suggestions for learners according to their preferences. Recommender systems are software tools that provide useful suggestions to learners by considering various types of knowledge and data, tracing learners’ profile, preferences, contextual information, and their interaction with the application. Personalized learning is an educational approach that customizes learning based on the learner’s characteristics.
In this thesis, an Innovative Model for constructing Personalized Learning Scenarios (IMPLS) was proposed. IMPLS is intended to assist educational institutions in designing personalized learning environments and to raise the awareness of instructors and learners on the importance of a personalized learning environment. The study validates the model by measuring its reliability based on Cronbach’s alphareliability test. Then developed a framework for ‘Recommending a Personalized Learning Subject” (RPLS) which provide an intelligent solution through suggesting the relevant and useful online learning subject to learners through utilizing filtering techniques.
In this research, a Recommendation-based Personalized Learning Subject (RPLS) model was implemented to recommend qualified learning subjects to learners. The RPLS model is based on a framework of learners’ characteristics and suggests suitable subjects to learners according to the Kolb’s experimental learning styles model. The RPLS model utilizes two machine-learning techniques, namely Decision Tree (DT) and K-Nearest Neighbor (KNN), to recommend personalized online learning subjects to learners. The KNN technique achieved an accuracy of 89.21% with a standard deviation ±6.98%, while the DTtechnique achieved an accuracy of 88.92% with a standard deviation ±7.27
The implemented RPLS model is able to provide learners with the suitable learning subject according to their characteristics through utilizing DT and KNN techniques. RPLSadopted various learning strategies through providing learners with the most suitable learning subject, which combines subject model learner model recommender model.