Ziegler, Cai-Nicolas, and Georg Lausen. "Analyzing correlation between trust and user similarity in online communities." International Conference on Trust Management. Berlin, Heidelberg: Springer Berlin Heidelberg, 2004.

This paper discusses the limitations of traditional collaborative filtering in recommender systems, especially in scenarios with distributed data storage and decentralized control, such as the Semantic Web. It proposes the use of computational trust models as a potential solution or supplement to enhance scalability and effectiveness. The study presents empirical results from an operational community focused on book recommendations, testing the hypothesis that trust metrics can reflect user similarity and improve recommender system outcomes.