Power-law weighted networks from local attachments

TL;DR

This paper presents a distributed method for constructing power-law weighted networks with specific properties, useful for modeling social and engineering systems, and demonstrates its effectiveness with real-world data.

Contribution

Introduces a novel distributed mechanism for creating power-law weighted networks with controllable scaling and clustering, applicable to empirical citation, patent, and legal opinion networks.

Findings

Connectivity distribution follows power-law behavior.

Networks resemble empirical citation and legal opinion distributions.

Framework enables modeling of complex real-world networks.

Abstract

This letter introduces a mechanism for constructing, through a process of distributed decision-making, substrates for the study of collective dynamics on extended power-law weighted networks with both a desired scaling exponent and a fixed clustering coefficient. The analytical results show that the connectivity distribution converges to the scaling behavior often found in social and engineering systems. To illustrate the approach of the proposed framework we generate network substrates that resemble steady state properties of the empirical citation distributions of (i) publications indexed by the Institute for Scientific Information from 1981 to 1997; (ii) patents granted by the U.S. Patent and Trademark Office from 1975 to 1999; and (iii) opinions written by the Supreme Court and the cases they cite from 1754 to 2002.