Parameterized Centrality for Network Analysis

TL;DR

This paper introduces a parameterized centrality measure based on Bonacich centrality for network analysis, enabling improved detection of important nodes and community structures by tuning the interaction length scale.

Contribution

It extends community detection methods to incorporate Bonacich centrality with a tunable parameter, offering a more flexible analysis of network structure.

Findings

Better identification of globally important nodes.

Enhanced community detection insights.

Application to benchmark networks shows improved results.

Abstract

Bonacich centrality measures the number of attenuated paths between nodes in a network. We use this metric to study network structure, specifically, to rank nodes and find community structure of the network. To this end we extend the modularity-maximization method for community detection to use this centrality metric as a measure of node connectivity. Bonacich centrality contains a tunable parameter that sets the length scale of interactions. By studying how rankings and discovered communities change when this parameter is varied allows us to identify globally important nodes and structures. We apply the proposed method to several benchmark networks and show that it leads to better insight into network structure than earlier methods.