A new method called hierarchical K-means combines K-means and hierarchical clustering to improve the efficiency of analyzing microarray data. By using a mix of divisive and agglomerative approaches, this method creates clusters more effectively than traditional methods. The results show that hierarchical K-means outperforms both hierarchical clustering in cluster quality and K-means clustering in computational speed. This new method provides a better way to group and analyze data from microarray experiments.