What is the most common reason for a k-means clustering algorithm to returns a sub-optimal
clustering of its input?
A.
Non-negative values for the distance function
B.
Input data set is too large
C.
Non-normal distribution of the input data
D.
Poor selection of the initial controls