A Fast Algorithm for the Discrete Core/Periphery Bipartitioning Problem

TL;DR

This paper introduces a new efficient and exact algorithm for the discrete core/periphery bipartitioning problem, capable of handling large networks quickly and identifying the core as the set of highest-degree actors.

Contribution

The paper presents a novel algorithm that guarantees optimal solutions for large networks, improving over previous methods that were limited in size or optimality.

Findings

Algorithm computes optimal partitioning in under a second for networks with thousands of actors.

Optimal core corresponds to actors with the highest degrees in the network.

Method outperforms existing approaches in efficiency and accuracy.

Abstract

Various methods have been proposed in the literature to determine an optimal partitioning of the set of actors in a network into core and periphery subsets. However, these methods either work only for relatively small input sizes, or do not guarantee an optimal answer. In this paper, we propose a new algorithm to solve this problem. This algorithm is efficient and exact, allowing the optimal partitioning for networks of several thousand actors to be computed in under a second. We also show that the optimal core can be characterized as a set containing the actors with the highest degrees in the original network.