MetaCluster: An open-source Python library for metaheuristic-based clustering problems

Clustering, based on metaheuristic algorithms, is a rapidly developing field. Its goal is to use these methods to reframe clustering issues as optimization problems. In this study, we propose an open-source library named MetaCluster. This library leverages the latest metaheuristic algorithms to tack...

Full description

Saved in:
Bibliographic Details
Published inSoftwareX Vol. 24; p. 101597
Main Authors Van Thieu, Nguyen, Oliva, Diego, Pérez-Cisneros, Marco
Format Journal Article
LanguageEnglish
Published Elsevier 01.12.2023
Subjects
Clustering
Evolutionary computation
Metaheuristic algorithms
Optimization
Online AccessGet full text
ISSN2352-7110
2352-7110
DOI10.1016/j.softx.2023.101597

Cover

More Information
Summary:Clustering, based on metaheuristic algorithms, is a rapidly developing field. Its goal is to use these methods to reframe clustering issues as optimization problems. In this study, we propose an open-source library named MetaCluster. This library leverages the latest metaheuristic algorithms to tackle partitional clustering challenges. MetaCluster has two primary goals: to be open-source and to provide a user-friendly interface with comprehensive documentation. This design aims to facilitate usage for individuals with varying experience levels across diverse applications and domains. The current version of MetaCluster is easy to install and deploy. It offers access to 48 available datasets, over 40 performance metrics, more than 40 evaluation measurement methods, and a wide array of 200 metaheuristic algorithms.
ISSN:2352-7110
2352-7110
DOI:10.1016/j.softx.2023.101597