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.
K. Behrenbruch, S. Jandt, L. Schmidt, and A. Roßnagel. Gestaltung nachhaltiger Arbeitssysteme - Wege zur gesunden, effizienten und sicheren Arbeit: 58. Kongress der Gesellschaft für Arbeitswissenschaft (Kassel 2012), page 473–476. Dortmund, GfA-Press, (2012)
L. Schmidt, L. Cramar, J. Hegenberg, R. Braun, and R. Herrmann. Human Machine Interaction Design: Gezielt wahrnehmen, sicher erkennen, attraktiv gestalten (Stuttgart 2013), page 27–36. Stuttgart, Institut für Konstruktionstechnik und Technisches Design der Universität Stuttgart, (2013)
L. Schmidt, and K. Behrenbruch. Grundlagen und Anwendungen der Mensch-Maschine-Interaktion: 10. Berliner Werkstatt Mensch-Maschine-Systeme (Berlin 2013), page 102–107. Berlin, Universitätsverlag der TU Berlin, (2013)
M. Radziwill, R. Kniewel, and L. Schmidt. International Scientific Conference on Mobility and Transport: ITS for Connected Mobility (München 2013), page 1–10. München, Institut für Verkehrswesen der Technischen Universität München, (2013)
R. Kniewel, C. Evers, L. Schmidt, and K. Geihs. Electronic Communications of the EASST : Proceedings of the Combined Workshop on Self-organizing, Adaptive, and Context- Sensitive Distributed Systems and Self-organized Communication in Disaster Scenarios (Stuttgart 2013), (2013)
M. Domhardt, E. Tunca, I. Zoller, P. Lotz, and L. Schmidt. Grundlagen und Anwendungen der Mensch-Maschine-Interaktion: 10. Berliner Werkstatt Mensch-Maschine-Systeme (Berlin 2013), page 9–18. Berlin, Universitätsverlag der TU Berlin, (2013)
M. Domhardt, and L. Schmidt. Chancen durch Arbeits-, Produkt- und Systemgestaltung - Zukunftsfähigkeit für Produktions- und Dienstleistungsunternehmen: 59. Kongress der Gesellschaft für Arbeitswissenschaft (Krefeld 2013), page 665–668. Dortmund, GfA-Press, (2013)
M. Domhardt, and L. Schmidt. Chancen durch Arbeits-, Produkt- und Systemgestaltung - Zukunftsfähigkeit für Produktions- und Dienstleistungsunternehmen: 59. Kongress der Gesellschaft für Arbeitswissenschaft (Krefeld 2013), page 229–232. Dortmund, GfA-Press, (2013)
R. Braun, and L. Schmidt. Lebensqualität im Wandel von Demografie und Technik: 6. Deutscher AAL-Kongress (Berlin 2013), page 321–325. Berlin, VDE-Verlag, (2013)
R. Braun, J. Hegenberg, L. Cramar, and L. Schmidt. Chancen durch Arbeits-, Produkt- und Systemgestaltung - Zukunftsfähigkeit für Produktions- und Dienstleistungsunternehmen: 59. Kongress der Gesellschaft für Arbeitswissenschaft (Krefeld 2013), page 413–416. Dortmund, GfA-Press, (2013)