A tabu-search-based heuristic for clustering

This paper considers a clustering problem where a given data set is partiti oned into a certain number of natural and homogeneous subsets such that eac h subset is composed of elements similar to one another but different from those of any other subset. For the clustering problem, a heuristic algorith m is exploited by combining the tabu search heuristic with two complementar y functional procedures, called packing and releasing procedures. The algor ithm is numerically tested for its effectiveness in comparison with referen ce works including the tabu search algorithm, the K-means algorithm and the simulated annealing algorithm. (C) 2000 Pattern Recognition Society. Publi shed by Elsevier Science Ltd. All rights reserved.