A novel clustering approach using hierarchical genetic algorithms

Clustering is a central task for data analysis that partitions heterogeneous data sets into groups of more homogeneous characteristics. However, most of clustering algorithms require the user to provide the number of clusters as input. In this paper, we consider the automatic clustering problem that one has to partition data points without any a priori knowledge about the correct number of clusters. The hierarchical genetic algorithm (HGA) is employed for automatically searching the number of clusters as well as properly locating the centers for clusters. The well-known Davies-Bouldin index is adopted as a measure of the validity of the clusters. Experimental results on artificial and real-life data sets are given to illustrate the effectiveness of the proposed approach.