A Comprehensive Study On Personalized Learning Recommendation In E-Learning System

Abstract

The Internet and cloud computing technology have enabled learners to choose courses based on their interests through e-learning systems. E-learning systems such as Massive Open Online Courses (MOOC) offer a comprehensive curriculum and teaching resources, including courseware, teaching videos, exercises, and homework. These systems provide free courses, rich content, and flexible selection. However, the abundance of teaching resources in e-learning systems can lead to information overload, making it challenging for learners to select suitable courses and resources. Personalized learning recommendation is a research field within intelligent learning. Its goal is to automatically and efficiently identify learners' characteristics and recommend matching learning resources to specific learners on e-learning systems to enhance learning motivation and effectiveness. This study examines the research articles on personalized learning recommendation technology and methodology published between 2013 and 2023, and only English articles and conference papers were selected. This study collects articles from five scientific databases: ACM Digital Library, IEEE Xplore, ScienceDirect, SpringerLink, and Worldwide Science. Out of 3413 identified articles, 64 relevant studies were selected for further systematic literature research. Only those with specific recommendation methods or implementation codes are chosen to ensure the quality of the articles. It summarizes the modeling of learners and learning objects and the algorithms used for personalized learning recommendations. Finally, the problems of current personalized learning recommendation methods are outlined, and views on future research opportunities are proposed.