Computational experience on four algorithms for the hard clustering problem

Khaled S. Al-Sultan*, M. Maroof Khan

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

73 Scopus citations

Abstract

In this paper, we consider the problem of clustering m objects in c clusters. The objects are represented by points in n-dimensional Euclidean space, and the objective is to classify these m points into c clusters such that the distance between points within a cluster and its center is minimized. The problem is a difficult optimization problem due to the fact that it posseses many local minima. Several algorithms have been developed to solve this problem which include the k-means algorithm, the simulated annealing algorithm, the tabu search algorithm, and the genetic algorithm. In this paper, we study the four algorithms and compare their computational performance for the clustering problem. We test these algorithms on several clustering problems from the literature as well as several random problems and we report on our computational experience.

Original languageEnglish
Pages (from-to)295-308
Number of pages14
JournalPattern Recognition Letters
Volume17
Issue number3
DOIs
StatePublished - 6 Mar 1996

Bibliographical note

Funding Information:
The authors are thankful to three anynomous referees whose comments have improved the quality of the paper. The support provided by Systems Engineering Department, King Fahd University of Petroleum and Minerals for conducting this research is greatly acknowledged.

Keywords

  • Clustering problem
  • Computational comparison
  • Genetic algorithm
  • Simulated annealing
  • Tabu search
  • k-means algorithm

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Computer Vision and Pattern Recognition
  • Artificial Intelligence

Fingerprint

Dive into the research topics of 'Computational experience on four algorithms for the hard clustering problem'. Together they form a unique fingerprint.

Cite this