Identifying hidden coalitions in the US House of Representatives by optimally partitioning signed networks based on generalized balance

TL;DR

This paper develops and applies advanced optimization models to identify hidden political coalitions in the US House of Representatives by partitioning signed networks, revealing a third coalition of effective legislators aligned with the majority.

Contribution

It introduces generalized binary linear programming models that guarantee globally optimal partitions of large signed networks, specifically applied to political collaboration networks.

Findings

A three-cluster partition outperforms the traditional two-cluster model.

The third coalition consists of highly effective legislators aligned with the majority.

Models are practical for networks with up to 30,000 edges.

Abstract

In network science, identifying optimal partitions of a signed network into internally cohesive and mutually divisive clusters based on generalized balance theory is computationally challenging. We reformulate and generalize two binary linear programming models that tackle this challenge, demonstrating their practicality by applying them them to partition networks of collaboration in the US House of Representatives. These models guarantee a globally optimal network partition and can be practically applied to signed networks containing up to 30,000 edges. In the US House context, we find that a three-cluster partition is better than a conventional two-cluster partition, where the otherwise hidden third coalition is composed of highly effective legislators who are ideologically aligned with the majority party.