Machine Learning (ML) K means vs K means++ The main difference between K Means and K Means++ lies in the selection of the centroids around which the clustering takes place and k means++ removes the drawback of K means which is it is dependent on initialization of centroid.