Hot-Get-Richer Network Growth Model

TL;DR

This paper introduces a 'hot-get-richer' network growth model that emphasizes recent degree changes to improve late node prominence, producing networks similar to preferential attachment models.

Contribution

The paper proposes a novel dynamical model incorporating recent degree change bias, enhancing late node ranking and network structure similarity to PA networks.

Findings

Produces later high-ranking nodes than PA model

Generates networks with structures similar to PA networks under certain parameters

Enhances late node prominence through recent degree change bias

Abstract

Under preferential attachment (PA) network growth models late arrivals are at a disadvantage with regard to their final degrees. Previous extensions of PA have addressed this deficiency by either adding the notion of node fitness to PA, usually drawn from some fitness score distributions, or by using fitness alone to control attachment. Here we introduce a new dynamical approach to address late arrivals by adding a recent-degree-change bias to PA so that nodes with higher relative degree change in temporal proximity to an arriving node get an attachment probability boost. In other words, if PA describes a rich-get-richer mechanism, and fitness-based approaches describe good-get-richer mechanisms, then our model can be characterized as a hot-get-richer mechanism, where hotness is determined by the rate of degree change over some recent past. The proposed model produces much later high-ranking nodes than the PA model and, under certain parameters, produces networks with structure similar to PA networks.