Bases: Algorithm
Assortativity computes a coefficient that expresses the correlation of a node attribute among connected pairs of nodes.
For a partition of the node set (a nominal attribute), the coefficient is computed from the mixing matrix e, where e_ij is the fraction of edge counts from nodes in subset i to nodes in subset j. In undirected graphs, each edge {u, v} with u != v is counted once in each direction and each self-loop is counted once, and the mixing matrix is normalized by the total of these counts (2m minus the number of self-loops). In directed graphs, each edge is counted once and the matrix is normalized by m.
G (networkit.graph) – The input graph.
data (list(float) or networkit.Partition) – Numerical node value array, or a partition of the node set.
Return the assortativity coefficient.
The assortativity coefficient.
float