An enhanced Walrus Optimizer with opposition-based learning and mutation strategy for data clustering

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F61989100%3A27240%2F25%3A10260604" target="_blank" >RIV/61989100:27240/25:10260604 - isvavai.cz</a>

Result on the web

<a href="https://www.sciencedirect.com/science/article/pii/S2590005625000360?pes=vor&utm_source=clarivate&getft_integrator=clarivate" target="_blank" >https://www.sciencedirect.com/science/article/pii/S2590005625000360?pes=vor&utm_source=clarivate&getft_integrator=clarivate</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1016/j.array.2025.100409" target="_blank" >10.1016/j.array.2025.100409</a>

Result language

angličtina

Original language name

An enhanced Walrus Optimizer with opposition-based learning and mutation strategy for data clustering

Original language description

Data clustering plays a crucial role in various domains, such as image processing, pattern recognition, and data mining. Traditional clustering techniques often suffer from limitations like sensitivity to initialization, poor convergence, and entrapment in local optima. To address these challenges, this paper proposes an Enhanced Walrus Optimizer (IWO) tailored for clustering tasks. The proposed IWO integrates two powerful strategies-Opposition-Based Learning (OBL) and Mutation Search Strategy (MSS)-to improve population diversity and prevent premature convergence, thereby enhancing both exploration and exploitation capabilities. These enhancements enable more accurate and stable identification of cluster centers. The effectiveness of IWO is validated through extensive experiments on multiple benchmark clustering datasets and compared against several state-of-the-art metaheuristic algorithms, including PSO, GWO, AOA, and others. The results demonstrate that IWO achieves better results, indicating improved compactness and separation of clusters. Statistical validation using p-values and ranking scores further confirms the superiority of the proposed method. These findings suggest that IWO offers a robust and flexible framework for solving complex clustering problems. Future work will explore hybrid deep learning-integrated models and parallel implementations to enhance scalability.

Czech name

—

Czech description

—

Type

J_imp - Article in a specialist periodical, which is included in the Web of Science database

CEP classification

—

OECD FORD branch

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Project

—

Continuities

O - Projekt operacniho programu

Publication year

2025

Confidentiality

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Name of the periodical

Array

ISSN

2590-0056

e-ISSN

—

Volume of the periodical

26

Issue of the periodical within the volume

JUL

Country of publishing house

NL - THE KINGDOM OF THE NETHERLANDS

Number of pages

26

Pages from-to

nestránkováno

UT code for WoS article

001499341900002

EID of the result in the Scopus database

—