Genetic algorithms applied to clustering problem and data mining

Clustering techniques have obtained adequate results when are applied to data mining problems. However, different runs of the same clustering technique on a specific dataset may result in different solutions. The cause of this difference is the choice of the initial cluster setting and the values of the parameters associated with the technique. A definition of good initial settings and optimal parameters values is not an easy task, particularly because both vary largely from one dataset to another. In this paper the authors investigate the use of Genetic Algorithms to determine the best initialization of clusters, as well as the optimization of the initial parameters. The experimental results show the great potential of the Genetic Algorithms for the improvement of the clusters, since they do not only optimize the clusters, but resolve the problem of the number K cluster, which had been giving it form a priori. The techniques of clustering are most used in the analysis of information or Data Mining, this method was applied to Data Set at mining.