The paper explores a novel ontological approach to user profiling in recommender systems, specifically for recommending online academic research papers. Two experimental systems, Quickstep and Foxtrot, are developed to create user profiles from unobtrusively monitored behavior and relevance feedback, using a research paper topic ontology. The systems use ontological inference to improve profiling accuracy and profile visualization to enhance feedback. Two small-scale experiments with 24 subjects over 3 months and a large-scale experiment with 260 subjects over an academic year are conducted to evaluate the approach. The results show that ontological inference improves user profiling, and external ontological knowledge can successfully bootstrap a recommender system. Profile visualization also improves profiling accuracy. The overall performance of the ontological recommender systems is favorable compared to other systems in the literature.The paper explores a novel ontological approach to user profiling in recommender systems, specifically for recommending online academic research papers. Two experimental systems, Quickstep and Foxtrot, are developed to create user profiles from unobtrusively monitored behavior and relevance feedback, using a research paper topic ontology. The systems use ontological inference to improve profiling accuracy and profile visualization to enhance feedback. Two small-scale experiments with 24 subjects over 3 months and a large-scale experiment with 260 subjects over an academic year are conducted to evaluate the approach. The results show that ontological inference improves user profiling, and external ontological knowledge can successfully bootstrap a recommender system. Profile visualization also improves profiling accuracy. The overall performance of the ontological recommender systems is favorable compared to other systems in the literature.