Algorithm 3: K-Means clustering

Input:x1…xl—objects, K—number of clusters

Output: {Ci, i =1 ..K}—cluster sets

generate at random K centroids c1, …, cK

while not converged do

Ci = x: ∥ x − ci∥2 ≤ minj=1..K ∥ x − cj∥2

$c_i=\frac{1}{\left|c_i\right|}{\sum }_{x\in C_i}x$

return{Ci}