An improved Branch-and-cut code for the maximum balanced subgraph of a signed graph

TL;DR

This paper presents an enhanced branch-and-cut algorithm for solving the Maximum Balanced Subgraph Problem in signed graphs, demonstrating improved computational performance on various real-world and random instances.

Contribution

The paper introduces a novel, improved branch-and-cut method specifically designed for the MBSP, advancing exact solution techniques for this problem.

Findings

Significant reduction in computational time on benchmark instances

Effective handling of diverse application-specific and random instances

Demonstrated superiority over previous algorithms

Abstract

The Maximum Balanced Subgraph Problem (MBSP) is the problem of finding a subgraph of a signed graph that is balanced and maximizes the cardinality of its vertex set. We are interested in the exact solution of the problem: an improved version of a branch-and-cut algorithm is proposed. Extensive computational experiments are carried out on a set of instances from three applications previously discussed in the literature as well as on a set of random instances.