Centrality in dynamic competition networks

TL;DR

This paper introduces a centrality measure based on the Dynamic Competition Hypothesis, predicting influential actors in competition networks by their common out-neighbors, and validates it across various real-world networks.

Contribution

It proposes a novel centrality score derived from the Dynamic Competition Hypothesis and demonstrates its effectiveness in identifying important actors in diverse networks.

Findings

The centrality measure predicts influential actors effectively.

It applies successfully to food webs, conflict networks, and voting data.

The measure outperforms some existing centrality metrics.

Abstract

Competition networks are formed via adversarial interactions between actors. The Dynamic Competition Hypothesis predicts that influential actors in competition networks should have a large number of common out-neighbors with many other nodes. We empirically study this idea as a centrality score and find the measure predictive of importance in several real-world networks including food webs, conflict networks, and voting data from Survivor.