Fig. 4
From: Modelling sequences and temporal networks with dynamic community structures
Inferred temporal model for a high school proximity network42. a The static part of the model divides the high school students into C = 10 groups (square nodes) that almost match the known classes (text labels). b The dynamic part of the model divides the directed multigraph group pairs in a into B N = B M = 9 groups (grey circles). The model corresponds to an n = 1 unified Markov chain on the edge labels, where the memory and tokens have identical partitions, as described in detail in the Methods section The unified first-order model
