Minimum Selective Subset on Some Graph Classes

TL;DR

This paper studies the Minimum Selective Subset problem in graphs, providing approximation algorithms, complexity results, and polynomial solutions for specific graph classes, advancing understanding of this NP-complete problem.

Contribution

It introduces a log(n)-approximation algorithm for general graphs, proves NP-completeness in planar graphs with two colors, and offers polynomial algorithms for trees and unit interval graphs.

Findings

Log(n)-approximation algorithm for general graphs

NP-completeness in planar graphs with two colors

Polynomial algorithms for trees and unit interval graphs

Abstract

In a connected simple graph G = (V(G),E(G)), each vertex is assigned a color from the set of colors C={1, 2,..., c}. The set of vertices V(G) is partitioned as V_1, V_2, ... ,V_c, where all vertices in V_j share the same color j. A subset S of V(G) is called Selective Subset if, for every vertex v in V(G), and if v is in V_j, at least one of its nearest neighbors in (S union (V(G)\ V_j)) has the same color as v. The Minimum Selective Subset (MSS) problem seeks to find a selective subset of minimum size. The problem was first introduced by Wilfong in 1991 for a set of points in the Euclidean plane, where two major problems, MCS (Minimum Consistent Subset) and MSS, were proposed. In graph algorithms, the only known result is that the MSS problem is NP-complete, as shown in 2018. Beyond this, no further progress has been made to date. In contrast, the MCS problem has been widely studied in various graph classes over the years. Therefore, in this work, we also extend the algorithmic study of MSS on various graph classes. We first present a log(n)-approximation algorithm for general graphs with n vertices and regardless of the number of colors. We also show that the problem remains NP-complete in planar graphs when restricted to just two colors.. Finally, we provide polynomial-time algorithms for computing optimal solutions in trees and unit interval graphs for any number of colors.