Hypergraph model of social tagging networks

TL;DR

This paper introduces an evolutionary hypergraph model for social tagging networks that captures key statistical properties and reveals insights into user behavior and resource retrieval in folksonomies.

Contribution

It presents a novel hypergraph model incorporating tag assignment and resource retrieval, aligning well with real-world data and enhancing understanding of folksonomy structures.

Findings

Model reproduces power-law hyperdegree distributions

Shows negative correlation between clustering and hyperdegree

Indicates tags facilitate resource retrieval and social connections

Abstract

The past few years have witnessed the great success of a new family of paradigms, so-called folksonomy, which allows users to freely associate tags to resources and efficiently manage them. In order to uncover the underlying structures and user behaviors in folksonomy, in this paper, we propose an evolutionary hypergrah model to explain the emerging statistical properties. The present model introduces a novel mechanism that one can not only assign tags to resources, but also retrieve resources via collaborative tags. We then compare the model with a real-world dataset: \emph{Del.icio.us}. Indeed, the present model shows considerable agreement with the empirical data in following aspects: power-law hyperdegree distributions, negtive correlation between clustering coefficients and hyperdegrees, and small average distances. Furthermore, the model indicates that most tagging behaviors are motivated by labeling tags to resources, and tags play a significant role in effectively retrieving interesting resources and making acquaintance with congenial friends. The proposed model may shed some light on the in-depth understanding of the structure and function of folksonomy.