DYPSOKM: A Dynamic Union Of PSO And K-Means, A Better Cluster

DOI

10.2991/jimec-17.2017.5How to use a DOI?

Keywords

clustering; K-means; PSO; multiple swarms; amplitude limiting of speed

Abstract

As one of the most famous clustering algorithms, K-means is simple and effective but easily falls into local optimal solution. Aimed at this flaw, many methods including PSO had been applied to optimize K-means. As a typical swarm intelligence optimization algorithm, PSO(particle swarm optimization) has better global convergence and robustness. This paper will applies the PSO to optimize the K-means clustering algorithm based on the basic PSOKM. On the one hand, we initialize the particles using dichotomy K-means. On the other hand, this paper in return utilizes the feature of multiple swarms shown in k-means to build the multiple swarms PSO. In main computational details, we light weight the calculation of multiple-population in order to enhance computational efficiency. Meanwhile, the speed of particles will be limited as a certain way to improve the validity of algorithm. Finally, Experiment results of our algorithm shows better convergence and validity compared with other algorithms mentioned in this paper.

Copyright

© 2017, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Qin Ke
AU  - Liusheng Huang
AU  - Hongli Xu
PY  - 2017/10
DA  - 2017/10
TI  - DYPSOKM: A Dynamic Union Of PSO And K-Means, A Better Cluster
BT  - Proceedings of the 2017 2nd Joint International Information Technology, Mechanical and Electronic Engineering Conference (JIMEC 2017)
PB  - Atlantis Press
SP  - 22
EP  - 28
SN  - 2352-538X
UR  - https://doi.org/10.2991/jimec-17.2017.5
DO  - 10.2991/jimec-17.2017.5
ID  - Ke2017/10
ER  -