Mapping Twitter Topic Networks: From Polarized Crowds to Community Clusters

Posted on by Brandon Klein

The network types we have identified group together in pairs based on their key properties. Networks can vary in terms of their internal divisions, density, and the direction of their connections. The first two network types are opposites of one another in terms of division or unity; the Polarized Crowd type is divided while the Tight Crowd network is unified. The next pair of network types, Brand Clusters and Community Clusters, have large populations of isolates, but vary in terms of the density of clustered connections. The Brand Cluster network structure has small disconnected groups with many isolated participants, while the Community Cluster network structure has larger, more connected groups along with many isolates. The last two networks are inversions of one another: the Broadcast Network features many spokes pointing inward to a hub while the Support Network structure features a hub linking outward to many spokes. Each of these network types is described in detail below.