A model for generating tunable clustering coefficients independent of the number of nodes in scale free and random networks

TL;DR

This paper introduces a new growth model for probabilistic networks that can independently tune clustering coefficients regardless of network size and explains the inverse relationship between local clustering and node degree.

Contribution

The paper presents a novel growth model that accurately predicts tunable clustering coefficients and the inverse local clustering-degree relationship in both random and scale-free networks.

Findings

Model predicts tunable clustering coefficients independent of network size

Model explains inverse relationship between local clustering and node degree

Applicable to both random and scale-free networks

Abstract

Probabilistic networks display a wide range of high average clustering coefficients independent of the number of nodes in the network. In particular, the local clustering coefficient decreases with the degree of the subtending node in a complicated manner not explained by any current models. While a number of hypotheses have been proposed to explain some of these observed properties, there are no solvable models that explain them all. We propose a novel growth model for both random and scale free networks that is capable of predicting both tunable clustering coefficients independent of the network size, and the inverse relationship between the local clustering coefficient and node degree observed in most networks.