Attack Vulnerability of Complex Networks in Center-Based Strategies

TL;DR

This paper investigates the vulnerability of complex networks to center-based node removal attacks, introducing a novel median-based strategy that proves highly destructive in both synthetic and real-world networks.

Contribution

It proposes a new median-based attack strategy and compares its effectiveness with other center-based attacks on various network types.

Findings

Median-based attack is highly destructive.

Recalculated network information improves attack efficiency.

Center-based attacks significantly disrupt network connectivity.

Abstract

Central nodes are critical in establishing structural connectivity in a complex network. Attacking such nodes can create real havoc in a complex system. We propose attack strategies based on four types of centers, namely betweenness center, degree center, median, and center. We study the vulnerability of synthetic as well as real-world networks in these center-based attacks. These attacks are node-removal attacks which involve identifying the central node set and removing them from the network. We observed that the attacks based on recalculated network information are more efficient than ones based on initial network information. This work shows that the median-based attack, which is a novel strategy proposed in this work, is highly destructive in real as well as synthetic networks.