Multi-faceted Learning Paths Recommendation Via Semantic Linked Network

Abstract

Cognition overload is one of the major problems in current self-learning intelligent learning systems. Providing learners with the personalized learning path can effectively smooth over users’ learning disorientation. In this paper, we propose a multi-faceted recommendation framework that provides learners with personalized learning paths based on their different learning styles. Building the recommendation system mainly involves the following three steps: (1) analyze the influences of the learning style in different dimensions during the learning process, (2) automatically organize the Learning Objects (LOs) into a multi-faceted Semantic Linked Network (SLN) via self-organized rules, (3) recommend the learning path to the learner through a reasoning machine based on the constructed SLN. The experiments verify the efficiency of the proposed method.