H. Halpin, V. Robu, und H. Shepherd. Proceedings of the 16th nternational World Wide Web Conference (WWW'07), New York, NY, USA, ACM Press, (2007)
The debate within the Web community over the optimal meansby which to organize information often pits formalized classi�cationsagainst distributed collaborative tagging systems. A numberof questions remain unanswered, however, regarding the nature ofcollaborative tagging systems including whether coherent categorizationschemes can emerge from unsupervised tagging by users.This paper uses data from the social bookmarking site del.icio.us toexamine the dynamics of collaborative tagging systems. In particular,we examine whether the distribution of the frequency of useof tags for popular sites with a long history (many tags and manyusers) can be described by a power law distribution, often characteristicof what are considered complex systems. We produce agenerative model of collaborative tagging in order to understandthe basic dynamics behind tagging, including how a power law distributionof tags could arise. We empirically examine the tagginghistory of sites in order to determine how this distribution arisesover time and to determine the patterns prior to a stable distribution.Lastly, by focusing on the high-frequency tags of a site wherethe distribution of tags is a stabilized power law, we show how tagco-occurrence networks for a sample domain of tags can be usedto analyze the meaning of particular tags given their relationship toother tags.
