The Blogosphere at a Glance—Content-Based Structures Made Simple

A network representation based on a basic wordoverlap similarity measure between blogs is introduced. The simplicity of the representation renders it computationally tractable, transparent and insensitive to representation-dependent artifacts. Using Swedish blog data, we demonstrate that the representation, in spite of its simplicity, manages to capture important structural properties of the content in the blogosphere. First, blogs that treat similar subjects are organized in distinct network clusters. Second, the network is hierarchically organized as clusters in turn form higher-order clusters: a compound structure reminiscent of a blog taxonomy.