In k-means clustering 01) K should be very large O2) K is how many standard deviations the clusters are apart 3) The computer determines how many clusters known as k O4) The analyst decides on the number of clusters k