Metadata-based recommendation approach for ontologies: The usage of metadata along with user profiles

Abstract

During the last 2 decades, many recommendation systems were proposed to deal with the complexity of user preferences prediction, many models were used under one of the 3 general filtering approaches (Collaborative, Content-based and Hybrid approach). In the other hand, Ontologies were integrated, and used in data formal representation, to improve interoperability between different agents, and thus what could give better results for Recommendation systems assisted with Ontology. This thesis defines the basic knowledge related to three topics that are necessary to build a metadata based recommendation system, assisted with user profiles, and cites different works on recommendation systems based on ontologies, or on structured datasets.