Centrality in Dynamic Competition Networks

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.

An Evolving Network Model from Clique Extension

“…We investigate the clique profiles on the following datasets, obtained from [5] and [16]. The networks in the dataset in [16] are known as ego-networks; that is, networks of reasonable order sampled from a larger network.…”

“…Each graph is sampled from users of the same gender; male and female are selected in their sampling and form labels. ( 5) Survivor and Big Brother [5]. We consider datasets from two social game television shows: Survivor and Big Brother.…”

Winner does not take all: contrasting centrality in adversarial networks

“…An approach taken in [5,6] in the detection of leaders in adversarial networks is the common out-neighbour score (or CON score). For nodes u, v, w in a graph G, we define w to be a common out-neighbor of u and v if (u, w) and (v, w) are two directed edges in G. We let CON(u, v) be the number of common outneighbour of distinct nodes u and v, and define…”