Which statement about K-means clustering is true?
In K-means clustering, the initial centroids are sometimes randomly selected.
To be accurate, you want your centroids outside of the cluster.
The number of clusters are always randomly selected.
K-means clustering is often used in supervised machine learning.