A Extension to k-Means and (k, p)-Clustering