Modified k-means algorithm and genetic approach for cluster optimization

Hierarchical clustering is of enormous importance in data analytics especially because of the exponential growth of the real world data. Frequently these data are unlabelled and there is small prior domain knowledge offered. In this work the plan is to improve the efficiency by introducing a set of methods dealt with synthetic and real data on agglomerative hierarchical clustering followed by k-means. Instead of building cluster hierarchies based on uncooked data points, and this approach builds a hierarchy based on a set of centroid assigned with the support of k-means. K-means algorithm with genetic approach for clustering is the new term and produce optimized results with large real world datasets are analyzed in this work.