Mechanisms for identifying clusters in graphs. Where these clusters define disjoint sets of vertices,
they may be used to define a VertexPartition
for more convenient manipulation of the vertex/set
relationships.
Current clustering algorithms include:
BicomponentClusterer
: finds all subsets of vertices for which at least
2 vertices must be removed in order to disconnect the induced subgraphs.
EdgeBetweennessClusterer
: identifies vertex clusters by removing the edges of the highest
'betweenness' scores (see the importance/scoring package).
VoltageClusterer
: Clusters vertices based on their ranks as
calculated by VoltageRanker
.
WeakComponentVertexClusterer
: Clusters vertices based on their membership in weakly
connected components of a graph.