A hypergraph model of social tagging networks

Abstract

The past few years have witnessed the great success of a new family of paradigms, so-calledfolksonomy, which allows users to freely associate tags with resources and efficientlymanage them. In order to uncover the underlying structures and user behaviors infolksonomy, in this paper, we propose an evolutionary hypergraph model for explaining theemerging statistical properties. The present model introduces a novel mechanism that cannot only assign tags to resources, but also retrieve resources via collaborativetags. We then compare the model with a real-world data set: Del.icio.us. Indeed,the present model shows considerable agreement with the empirical data in thefollowing aspects: power-law hyperdegree distributions, negative correlation betweenclustering coefficients and hyperdegrees, and small average distances. Furthermore, themodel indicates that most tagging behaviors are motivated by labeling tags onresources, and the tag plays a significant role in effectively retrieving interestingresources and making acquaintances with congenial friends. The proposed model mayshed some light on the in-depth understanding of the structure and function offolksonomy.