Social bookmarking systems and their emergent information structures, known as folksonomies, are increasingly important data sources for Semantic Web applications. A key question for harvesting semantics from these systems is how to extend and adapt traditional notions of similarity to folksonomies, and which measures are best suited for applications such as navigation support, semantic search, and ontology learning. Here we build an evaluation framework to compare various general folksonomy-based similarity measures derived from established information-theoretic, statistical, and practical measures. Our framework deals generally and symmetrically with users, tags, and resources. For evaluation purposes we focus on similarity among tags and resources, considering different ways to aggregate annotations across users. After comparing how tag similarity measures predict user-created tag relations, we provide an external grounding by user-validated semantic proxies based on WordNet and the Open Directory. We also investigate the issue of scalability. We find that mutual information with distributional micro-aggregation across users yields the highest accuracy, but is not scalable; per-user projection with collaborative aggregation provides the best scalable approach via incremental computations. The results are consistent across resource and tag similarity.
Social bookmarking systems and their emergent information structures, known as folksonomies, are increasingly important data sources for Semantic Web applications. A key question for harvesting semantics from these systems is how to extend and adapt traditional notions of similarity to folksonomies, and which measures are best suited for applications such as navigation support, semantic search, and ontology learning. Here we build an evaluation framework to compare various general folksonomy-based similarity measures derived from established information-theoretic, statistical, and practical measures. Our framework deals generally and symmetrically with users, tags, and resources. For evaluation purposes we focus on similarity among tags and resources, considering different ways to aggregate annotations across users. After comparing how tag similarity measures predict user-created tag relations, we provide an external grounding by user-validated semantic proxies based on WordNet and the Open Directory. We also investigate the issue of scalability. We find that mutual information with distributional micro-aggregation across users yields the highest accuracy, but is not scalable; per-user projection with collaborative aggregation provides the best scalable approach via incremental computations. The results are consistent across resource and tag similarity.
B. Klein, S. Lau, A. Pirali, T. Löffler, and K. David. ASWN '08: Proceedings of the 2008 Eighth International Workshop on Applications and Services in Wireless Networks, page 20--25. Washington, DC, USA, IEEE Computer Society, (2008)
R. Jäschke, B. Krause, A. Hotho, and G. Stumme. Proceedings of the Second International Conference on Weblogs and Social Media(ICWSM 2008), AAAI Press, (2008)
R. Jäschke, A. Hotho, C. Schmitz, B. Ganter, and G. Stumme. Proceedings of the 6th IEEE International Conference on Data Mining (ICDM 06), page 907-911. Hong Kong, IEEE Computer Society, (December 2006)
C. Schmitz, A. Hotho, R. Jäschke, and G. Stumme. Data Science and Classification: Proc. of the 10th IFCS Conf., page 261--270. Berlin, Heidelberg, Springer, (2006)
B. Krause, A. Hotho, and G. Stumme. Advances in Information Retrieval, 30th European Conference on IR Research, ECIR 2008, volume 4956 of LNAI, page 101-113. Heidelberg, Springer, (2008)
A. Hotho, R. J�schke, C. Schmitz, and G. Stumme. The Semantic Web: Research and Applications, volume 4011 of LNAI, page 411-426. Heidelberg, Springer, (June 2006)
M. Grahl, A. Hotho, and G. Stumme. 7th International Conference on Knowledge Management (I-KNOW '07), page 356-364. Graz, Austria, Know-Center, (September 2007)
R. Jäschke, A. Hotho, C. Schmitz, and G. Stumme. Proceedings of the 15th International Conference on Conceptual Structures (ICCS 2007), volume 4604 of Lecture Notes in Artificial Intelligence, page 283--295. Berlin, Heidelberg, Springer-Verlag, (July 2007)
R. Jäschke, L. Marinho, A. Hotho, L. Schmidt-Thieme, and G. Stumme. Knowledge Discovery in Databases: PKDD 2007, 11th European Conference on Principles and Practice of Knowledge Discovery in Databases, volume 4702 of Lecture Notes in Computer Science, page 506-514. Berlin, Heidelberg, Springer, (2007)
R. Jäschke, A. Hotho, C. Schmitz, and G. Stumme. Proc. 18. Workshop Grundlagen von Datenbanken, page 80-84. Halle-Wittenberg, Martin-Luther-Universität, (June 2006)
A. Hotho, R. Jäschke, C. Schmitz, and G. Stumme. Proc. First International Conference on Semantics And Digital Media Technology (SAMT), volume 4306 of LNCS, page 56-70. Heidelberg, Springer, (December 2006)
B. Berendt, A. Hotho, and G. Stumme. Proc. of the 1st Intl. Workshop on Representation and Analysis of Web Space, page 1-16. Technical University of Ostrava, (2005)
C. Schmitz, A. Hotho, R. Jäschke, and G. Stumme. The Semantic Web: Research and Applications, volume 4011 of LNAI, page 530-544. Heidelberg, Springer, (2006)
G. Stumme. Conceptual Structures at Work: 12th International Conference on Conceptual Structures (ICCS 2004), volume 3127 of LNCS, page 109-125. Heidelberg, Springer, (2004)
P. Cimiano, A. Hotho, G. Stumme, and J. Tane. Concept Lattices, volume 2961 of LNAI, page 189-207. Heidelberg, Second International Conference on Formal Concept Analysis, ICFCA 2004, Springer, (2004)
G. Stumme. Proc. 3rd Intl. Conf. on Formal Concept Analysis, volume 3403 of Lecture Notes in Computer Science, page 315-328. Heidelberg, Springer, (2005)