New Directions to Improve Clustering with a Scatter Search Based Algorithm

We introduce in this paper a new global search algorithm to solve clustering problems. This new proposal uses one of the well known evolutionary algorithms called Scatter Search. This algorithm operates on a small set of solutions and makes only a limited use of randomization for diversification when searching for globally optimal solutions. The proposed method determines the number of clusters and the cluster centres in such a way that locally optimal solutions are avoided. We have applied this algorithm to standard and real world data bases and we have obtained good results compared to the K-means and Antclass algorithms.