A Semantic Recommender System for Learning Based on Encyclopedia of Digital Publication

Abstract

Digital publication is a useful and authoritative resource for knowledge and learning. How to use the knowledge in digital publication resources so as to enhance learning is an interesting and important task. Most of the recommender systems use users' preferences or history data for computation, which cannot solve the problems such as cold start, scarcity of history data or preferences data. A semantic recommender system is presented in this paper based on encyclopedic knowledge from digital publication resources, without considering history data or preferences data for learning the knowledge of a specific domain. Semantic relatedness is computed between concepts from the encyclopedia. The related concepts are recommended to users when one concept is reviewed. The method shows potential usability for domain-specific knowledge service. © Springer International Publishing Switzerland 2014.