Differential evolution and particle swarm optimisation in partitional clustering

Many partitional clustering algorithms based on genetic algorithms (GA) have been proposed to tackle the problem of finding the optimal partition of a data set. Very few studies considered alternative stochastic search heuristics other than GAs or simulated annealing. Two promising algorithms for nu...

Full description

Saved in:
Bibliographic Details
Published in:Computational statistics & data analysis 2006-03, Vol.50 (5), p.1220-1247
Main Authors: Paterlini, Sandra, Krink, Thiemo
Format: Article
Language:English
Subjects:
Cluster analysis
Differential evolution
Exact sciences and technology
General topics
Genetic algorithms
Mathematics
Multivariate analysis
Numerical analysis
Numerical analysis. Scientific computation
Numerical methods in probability and statistics
Particle swarm optimisation
Partitional clustering
Probability and statistics
Sciences and techniques of general use
Statistics
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Many partitional clustering algorithms based on genetic algorithms (GA) have been proposed to tackle the problem of finding the optimal partition of a data set. Very few studies considered alternative stochastic search heuristics other than GAs or simulated annealing. Two promising algorithms for numerical optimisation, which are hardly known outside the search heuristics field, are particle swarm optimisation (PSO) and differential evolution (DE). The performance of GAs for a representative point evolution approach to clustering is compared with PSO and DE. The empirical results show that DE is clearly and consistently superior compared to GAs and PSO for hard clustering problems, both with respect to precision as well as robustness (reproducibility) of the results. Only for simple data sets, the GA and PSO can obtain the same quality of results. Apart from superior performance, DE is easy to implement and requires hardly any parameter tuning compared to substantial tuning for GAs and PSOs. Our study shows that DE rather than GAs should receive primary attention in partitional clustering algorithms.
ISSN:0167-9473
1872-7352
DOI:10.1016/j.csda.2004.12.004