The Semantic Recommender is the SONIC laboratory’s next-generation recommender system, based on semantic web-related techologies. The Semantic Recommender embraces the Multi-Theoretical, Multi-Level (MTML) framework of social drivers to capture the motivations of the seeker of a recommendation. The Semantic Recommender integrates social network analysis and automated reasoning approaches to making recommendations. The Semantic Recommender utilitizes Resource Description Framework (RDF) as a representation for instance data, Web Ontology Language (OWL) as a representation for the domain ontology, the Java Universal Network/Graph Framework (JUNG) for network analysis, and the Pellet reasoning engine.
A demonstration of the Semantic Recommender, operating in the domain of recommending collaborations among participants in the Northwestern University Clinical and Translational Sciences Institute (NUCATS) is available online. A demonstration script will help you get started.
The Semantic Recommender is a collaboration with Maryam Fazel-Zarandi, PhD candidate at the Department of Computer Science, University of Toronto, and Noshir Contractor, Yun Huang, and Hugh Devlin of the SONIC lab.