Characterizing the head of the degree distributions of growing networks

TL;DR

This paper introduces a comprehensive model for growing networks that combines mixed attachment mechanisms, reciprocity, and variable edge addition, providing analytical insights into the formation of degree distributions with extended tails.

Contribution

It extends existing models by incorporating variable edge addition distributions and reciprocity, offering analytical expressions for degree distribution evolution in growing networks.

Findings

Number of new edges influences degree distribution tail behavior.

Reciprocity mechanisms significantly affect the network's degree structure.

Analytical expressions accurately predict the evolution of in-degree distributions.

Abstract

The analysis in this paper helps to explain the formation of growing networks with degree distributions that follow extended exponential or power-law tails. We present a generic model in which edge dynamics are driven by a continuous attachment of new nodes and a mixed attachment mechanism that triggers random or preferential attachment. Furthermore, reciprocal edges to newly added nodes are established according to a response mechanism. The proposed framework extends previous mixed attachment models by allowing the number of new edges to vary according to various discrete probability distributions, including Poisson, Binomial, Zeta, and Log-Series. We derive analytical expressions for the limit in-degree distribution that results from the mixed attachment and response mechanisms. Moreover, we describe the evolution of the dynamics of the cumulative in-degree distribution. Simulation results illustrate how the number of new edges and the process of reciprocity significantly impact the head of the degree distribution.