Clustering is a popular data analysis and data mining technique. Among different proposed methods, k-means is an efficient clustering technique to cluster datasets, but this method highly depends on the initial state and usually converges to local optimum solution. This paper takes the advantage of a novel evolutionary algorithm, called artificial bee colony (ABC), to improve the capability of k-means in finding global optimum clusters in nonlinear partitional clustering problems. The proposed method is the combination of k-means and ABC algorithms, called kABC, which can find better cluster portions. Both kABC and k-means are run on three known data sets from the UCI Machine Learning Repository. The simulation results show that the combination of ABC and k-means technique has more ability to search for global optimum solutions and more ability for passing local optimum.

Clustering Analysis with Combination of Artificial Bee Colony Algorithm and k-Means Technique

ARMANO, GIULIANO;FARMANI, MOHAMMAD REZA
2014-01-01

Abstract

Clustering is a popular data analysis and data mining technique. Among different proposed methods, k-means is an efficient clustering technique to cluster datasets, but this method highly depends on the initial state and usually converges to local optimum solution. This paper takes the advantage of a novel evolutionary algorithm, called artificial bee colony (ABC), to improve the capability of k-means in finding global optimum clusters in nonlinear partitional clustering problems. The proposed method is the combination of k-means and ABC algorithms, called kABC, which can find better cluster portions. Both kABC and k-means are run on three known data sets from the UCI Machine Learning Repository. The simulation results show that the combination of ABC and k-means technique has more ability to search for global optimum solutions and more ability for passing local optimum.
2014
Artificial bee colony algorithm; k-means
File in questo prodotto:
File Dimensione Formato  
2014-IJCTE-armano.pdf

accesso aperto

Dimensione 269.18 kB
Formato Adobe PDF
269.18 kB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11584/134250
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact