Font Size: 

We present a model of network transformation mediated by confinement.
With a simple network dynamics and two parameters directly connected
with real-world quantities, the model has the ability to generate
complex structures similar to real-world networks. From any initial
network, the nodes are randomly selected to be temporarily confined.
The confined nodes form new links between them at the same rate that
they lose connections with the external nodes. As the network evolves,
a series of characteristics of non-trivial networks emerge: the
formation of stable distributions of heterogeneous degrees similar to
those of empirical networks, a growing clustering coefficient and the
emergence of communities outside the confined space.
Unlike the traditional benchmarks used to create modular networks,
there is no arbitrary determination of the number of modules,
nor node metadata that define it as a member of a particular community,
nor a tunable parameter directly related to the expected modularity.
The modules emerge as a result of the dynamics, while the nodes can move
between them as connections are rewired. The proposed algorithm has
the potential to simulate cases of community dynamics in situations
where time-stamped network data is scarce or absent and, at the same
time, provides a reasonable framework for studying the dynamics of
real networks.