Approximation Algorithms For The Dispersion Problems in a Metric Space

In this article, we consider the $c$-dispersion problem in a metric space $(X,d)$. Let $P=\{p_{1}, p_{2}, \ldots, p_{n}\}$ be a set of $n$ points in a metric space $(X,d)$. For each point $p \in P$ and $S \subseteq P$, we define $cost_{c}(p,S)$ as the sum of distances from $p$ to the nearest $c$ points in $S \setminus \{p\}$, where $c\geq 1$ is a fixed integer. We define $cost_{c}(S)=\min_{p \in S}\{cost_{c}(p,S)\}$ for $S \subseteq P$. In the $c$-dispersion problem, a set $P$ of $n$ points in a metric space $(X,d)$ and a positive integer $k \in [c+1,n]$ are given. The objective is to find a subset $S\subseteq P$ of size $k$ such that $cost_{c}(S)$ is maximized. We propose a simple polynomial time greedy algorithm that produces a $2c$-factor approximation result for the $c$-dispersion problem in a metric space. The best known result for the $c$-dispersion problem in the Euclidean metric space $(X,d)$ is $2c^2$, where $P \subseteq \mathbb{R}^2$ and the distance function is Euclidean distance [ Amano, K. and Nakano, S. I., Away from Rivals, CCCG, pp.68-71, 2018 ]. We also prove that the $c$-dispersion problem in a metric space is $W[1]$-hard.