Modified Particle Swarm Optimization for Clustering

Muchamad Kurniawan, Rani Muhima, Maftahatul Hakimah, Siti Agustini, Rahmi Putri

2023

Abstract

The traditional clustering analysis algorithm grouped into several types, one of the most popular is clustering based on partition. One of the limitations of partition clustering is that the initial centroid. initialization is critical. Previous studies have used optimization algorithms such as Particle Swarm Optimization (PSO) to obtain initial centroids. The first contribution in research is to use PSO with the addition of the Mean process to produce a clustering analysis we call it PSO Mean Clustering (PMC). The second contribution is to use a partial Gaussian distribution to generate the initial population in the PMC method, and we call it Gaussian PSO Mean Clustering (GPMC). The datasets used in this research are six clustering datasets to get an internal and external evaluation. The results obtained by the two proposed methods are better than the PSO clustering method and traditional K-means based on internal and external evaluation methods compared. Average value internal evaluation percentage of GPMC across K-means is 3.94%.

Paper Citation

in Harvard Style

Kurniawan M., Muhima R., Hakimah M., Agustini S. and Putri R. (2023). Modified Particle Swarm Optimization for Clustering. In Proceedings of the 4th International Conference on Advanced Engineering and Technology - Volume 1: ICATECH; ISBN 978-989-758-663-7, SciTePress, pages 89-96. DOI: 10.5220/0012111800003680

in Bibtex Style

@conference{icatech23,
author={Muchamad Kurniawan and Rani Muhima and Maftahatul Hakimah and Siti Agustini and Rahmi Putri},
title={Modified Particle Swarm Optimization for Clustering},
booktitle={Proceedings of the 4th International Conference on Advanced Engineering and Technology - Volume 1: ICATECH},
year={2023},
pages={89-96},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012111800003680},
isbn={978-989-758-663-7},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 4th International Conference on Advanced Engineering and Technology - Volume 1: ICATECH
TI - Modified Particle Swarm Optimization for Clustering
SN - 978-989-758-663-7
AU - Kurniawan M.
AU - Muhima R.
AU - Hakimah M.
AU - Agustini S.
AU - Putri R.
PY - 2023
SP - 89
EP - 96
DO - 10.5220/0012111800003680
PB - SciTePress