The Influences of Edge Asymmetry on Network Robustness

TL;DR

This paper introduces Edge Asymmetry (EA) as a new measure to analyze network robustness, demonstrating that EA-based attack strategies are effective and computationally efficient across various real-world networks.

Contribution

The paper proposes a novel Edge Asymmetry measure and develops an EA-based attack strategy that outperforms traditional methods in efficiency and effectiveness.

Findings

EA-based attack matches edge betweenness performance

EA attack outperforms edge degree and random attacks

Higher network asymmetry improves EA attack effectiveness

Abstract

Asymmetry of in/out-degree distribution is a widespread phenomenon in real-world complex networks. This paper put forward the concept of Edge Asymmetry(EA) to quantify this feature. We designed an EA-based strategy to attack six kinds of real-world networks and found that it was able to achieve the effect as well as edge betweenness-based(EB) and better than edge degree-based(ED) and random attack strategies. In simulation, we found that the greater the network asymmetry the better the EA-based attack strategy performed. By definition, the computational complexity of EA was much lower than that of EB. Therefore, EA-based attack strategies were superior in efficiency. We verified the effect of the EA-based attack strategy with four groups of large-scale networks.