Network Evolution by Relevance and Importance Preferential Attachment

TL;DR

This paper introduces a new network evolution model based on relevance and importance, capturing key real-world network features like scale-free distribution and small-world properties, and compares it with empirical data.

Contribution

It proposes a novel preferential attachment model incorporating relevance and importance, enhancing the understanding of real-world network formation.

Findings

The model produces scale-free degree distributions.

It exhibits high clustering coefficients.

It captures small-world and core-periphery structures.

Abstract

Relevance and importance are the main factors when humans build network connections. We propose an evolutionary network model based on preferential attachment(PA) considering these factors. We analyze and compute several important features of the network class generated by this algorithm including scale free degree distribution, high clustering coefficient, small world property and core-periphery structure. We then compare this model with other network models and empirical data such as inter-city road transportation and air traffic networks.