Scale Free Subnetworks by Design and Dynamics

TL;DR

This paper explores methods to create and analyze scale-free subnetworks through design and dynamic node exchange, comparing their properties with random and Barabási-Albert networks.

Contribution

It formalizes the transformation between subnetworks and weighted networks and introduces dynamic processes for generating scale-free subnetworks.

Findings

Designed subnetworks are scale-free and differ from random networks.

Dynamic node exchange can produce scale-free structures.

Hierarchical measurements distinguish different network types.

Abstract

This article addresses the degree distribution of subnetworks, namely the number of links between the nodes in each subnetwork and the remainder of the structure (cond-mat/0408076). The transformation from a subnetwork-partitioned model to a standard weighted network, as well as its inverse, are formalized. Such concepts are then considered in order to obtain scale free subnetworks through design or through a dynamics of node exchange. While the former approach allows the immediate derivation of scale free subnetworks, in the latter nodes are sequentially selected with uniform probability among the subnetworks and moved into another subnetwork with probability proportional to the degree of the latter. Comparison of the designed scale-free subnetworks with random and Barab\'asi-Albert counterparts are performed in terms of a set of hierarchical measurements.