MCS-HMS: A Multi-Cluster Selection Strategy for the Human Mental Search Algorithm

TL;DR

This paper introduces MCS-HMS, an improved Human Mental Search algorithm that enhances exploration and speed through multi-cluster selection and one-step k-means clustering, outperforming existing metaheuristics.

Contribution

The paper proposes a novel multi-cluster selection strategy and a faster clustering method to improve HMS's exploration ability and computational efficiency.

Findings

MCS-HMS outperforms HMS in optimization tasks.

MCS-HMS demonstrates superior exploration capabilities.

The one-step k-means speeds up the clustering phase.

Abstract

Population-based metaheuristic algorithms have received significant attention in global optimisation. Human Mental Search (HMS) is a relatively recent population-based metaheuristic that has been shown to work well in comparison to other algorithms. However, HMS is time-consuming and suffers from relatively poor exploration. Having clustered the candidate solutions, HMS selects a winner cluster with the best mean objective function. This is not necessarily the best criterion to choose the winner group and limits the exploration ability of the algorithm. In this paper, we propose an improvement to the HMS algorithm in which the best bids from multiple clusters are used to benefit from enhanced exploration. We also use a one-step k-means algorithm in the clustering phase to improve the speed of the algorithm. Our experimental results show that MCS-HMS outperforms HMS as well as other population-based metaheuristic algorithms