A general method for identifying node spreading influence via the adjacent matrix and spreading rate

TL;DR

This paper introduces a systematic method combining adjacency matrix and spreading rate to accurately identify influential nodes in complex networks, outperforming traditional centrality measures.

Contribution

The paper presents a novel approach that integrates network structure and spreading dynamics using eigenvector analysis to improve influence identification.

Findings

The method accurately predicts influential nodes in real networks.

It outperforms degree, K-shell, and eigenvector centrality measures.

The approach is validated against SIR epidemic simulations.

Abstract

With great theoretical and practical significance, identifying the node spreading influence of complex network is one of the most promising domains. So far, various topology-based centrality measures have been proposed to identify the node spreading influence in a network. However, the node spreading influence is a result of the interplay between the network topology structure and spreading dynamics. In this paper, we build up the systematic method by combining the network structure and spreading dynamics to identify the node spreading influence. By combining the adjacent matrix $A$ and spreading parameter $\beta$, we theoretical give the node spreading influence with the eigenvector of the largest eigenvalue. Comparing with the Susceptible-Infected-Recovered (SIR) model epidemic results for four real networks, our method could identify the node spreading influence more accurately than the ones generated by the degree, K-shell and eigenvector centrality. This work may provide a systematic method for identifying node spreading influence.