This paper presents an ontological approach to user profiling in recommender systems, focusing on the problem of recommending online academic research papers. Two experimental systems, Quickstep and Foxtrot, were developed to create user profiles from unobtrusively monitored behavior and relevance feedback, representing these profiles using a research paper topic ontology. A novel profile visualization approach was used to acquire profile feedback. Research papers were classified using ontological classes, and collaborative recommendation algorithms were used to recommend papers seen by similar people on their current topics of interest. Two small-scale experiments with 24 subjects over 3 months and a large-scale experiment with 260 subjects over an academic year were conducted to evaluate different aspects of the approach. Ontological inference was shown to improve user profiling, external ontological knowledge was used to successfully bootstrap a recommender system, and profile visualization was employed to improve profiling accuracy. The overall performance of the ontological recommender systems was presented and favorably compared to other systems in the literature. The paper discusses the problem domain and general approach to ontological recommendation, along with related work. It describes the Quickstep recommender system and an experiment to show how inference can improve user profiling and recommendation accuracy. It details an integration between the Quickstep system and an external ontology, along with an experiment to demonstrate its effectiveness at bootstrapping profiles. It describes the Foxtrot system and an experiment to demonstrate how profile visualization can be used to acquire feedback and improve profile accuracy. The paper concludes by bringing the work together, collating evidence supporting ontological user profiling in recommender systems and discussing future work. The paper presents an ontological approach to user profiling in recommender systems, using ontological inference and external ontologies to improve profiling accuracy and reduce the cold-start problem. The Quickstep system uses ontological inference to improve profiling accuracy and integrates an external ontology for profile bootstrapping. The Foxtrot system enhances Quickstep by employing profile visualization to acquire direct profile feedback. The paper evaluates the performance of these systems through experiments with 24 and 260 subjects, showing that ontological inference and external ontologies improve user profiling and recommendation accuracy. The results indicate that ontological approaches outperform traditional methods in terms of accuracy and robustness. The paper also discusses the integration of the Quickstep system with an external ontology, demonstrating how external ontologies can be used to bootstrap the recommender system and reduce the cold-start problem. The results of the experiments show that the new-system and new-user initial profiling algorithms improve profile accuracy and reduce the cold-start problem. The paper concludes that ontological approaches to user profiling in recommender systems offer significant improvements in accuracy and robustness, and that further research is needed to explore the potential of these approaches in larger-scale systems.This paper presents an ontological approach to user profiling in recommender systems, focusing on the problem of recommending online academic research papers. Two experimental systems, Quickstep and Foxtrot, were developed to create user profiles from unobtrusively monitored behavior and relevance feedback, representing these profiles using a research paper topic ontology. A novel profile visualization approach was used to acquire profile feedback. Research papers were classified using ontological classes, and collaborative recommendation algorithms were used to recommend papers seen by similar people on their current topics of interest. Two small-scale experiments with 24 subjects over 3 months and a large-scale experiment with 260 subjects over an academic year were conducted to evaluate different aspects of the approach. Ontological inference was shown to improve user profiling, external ontological knowledge was used to successfully bootstrap a recommender system, and profile visualization was employed to improve profiling accuracy. The overall performance of the ontological recommender systems was presented and favorably compared to other systems in the literature. The paper discusses the problem domain and general approach to ontological recommendation, along with related work. It describes the Quickstep recommender system and an experiment to show how inference can improve user profiling and recommendation accuracy. It details an integration between the Quickstep system and an external ontology, along with an experiment to demonstrate its effectiveness at bootstrapping profiles. It describes the Foxtrot system and an experiment to demonstrate how profile visualization can be used to acquire feedback and improve profile accuracy. The paper concludes by bringing the work together, collating evidence supporting ontological user profiling in recommender systems and discussing future work. The paper presents an ontological approach to user profiling in recommender systems, using ontological inference and external ontologies to improve profiling accuracy and reduce the cold-start problem. The Quickstep system uses ontological inference to improve profiling accuracy and integrates an external ontology for profile bootstrapping. The Foxtrot system enhances Quickstep by employing profile visualization to acquire direct profile feedback. The paper evaluates the performance of these systems through experiments with 24 and 260 subjects, showing that ontological inference and external ontologies improve user profiling and recommendation accuracy. The results indicate that ontological approaches outperform traditional methods in terms of accuracy and robustness. The paper also discusses the integration of the Quickstep system with an external ontology, demonstrating how external ontologies can be used to bootstrap the recommender system and reduce the cold-start problem. The results of the experiments show that the new-system and new-user initial profiling algorithms improve profile accuracy and reduce the cold-start problem. The paper concludes that ontological approaches to user profiling in recommender systems offer significant improvements in accuracy and robustness, and that further research is needed to explore the potential of these approaches in larger-scale systems.