K-Means Clustering Algorithm -- from Wolfram MathWorld

An algorithm for partitioning (or clustering) N data points into K disjoint subsets S_j containing N_j data points so as to minimize the sum-of-squares criterion J=sum_(j=1)^Ksum_(n in S_j)|x_n-mu_j|^2, where x_n is a vector representing the nth data poin

mathworld.wolfram.com