Interdependence of dynamical signals and topology: Detecting the influential nodes in networks

TL;DR

This paper investigates how the correlation of signals in various dynamical processes relates to network topology, proposing a method to identify influential nodes based on this relationship and validating its effectiveness.

Contribution

It introduces a novel method for detecting influential nodes in networks by analyzing the correlation between dynamical signals and node degree, validated across multiple systems.

Findings

Positive correlation between signal similarity and node degree

Method effectively identifies influential nodes

Potential applications in real-world systems

Abstract

By studying varies dynamical processes, including coupled maps, cellular automata and coupled differential equations, on five different kinds of known networks, we found a positive relation between signal correlation and node's degree. Thus a method of identifying influential nodes in dynamical systems is proposed, its validity is studied, and potential applications on real systems are discussed.