Exploiting Synergy Between Ontologies and Recommender Systems

Title: Exploiting Synergy Between Ontologies and Recommender Systems

Abstract: This research investigated the integration of recommender systems and ontologies to address the cold-start problem and interest-acquisition problem. The study used Quickstep, an academic domain ontology, and OntoCoPI to create a hybrid system. The results showed improved performance and user engagement, suggesting that the synergy between ontologies and recommender systems can enhance the effectiveness of both systems.

Main Research Question: How can the synergy between ontologies and recommender systems be exploited to improve the performance and user engagement of both systems?

Methodology: The study combined two existing systems: Quickstep, a web-based recommender system, and an ontology for the academic domain. OntoCoPI, a community of practice identifier, was used to match users with similar interests. The systems were integrated to share information, allowing the recommender system to learn from the ontology's knowledge and the ontology to benefit from the recommender system's dynamic user interests.

Results: The empirical evaluation of the integrated system demonstrated improved performance in terms of accuracy and efficiency. User engagement was also enhanced, as users found the system more intuitive and personalized.

Implications: The synergy between ontologies and recommender systems can lead to significant improvements in the performance and user engagement of both systems. This study provides a framework for integrating these two systems and suggests potential applications in various domains.

Link to Article: https://arxiv.org/abs/0204012v1 Authors: arXiv ID: 0204012v1