I'm working with the K-Means Algorithm in R and I want to figure out the differences of the 4 Algorithms Lloyd,Forgy,MacQueen and Hartigan-Wong which are available for the function "kmeans" in the stats package.
However I was notable to get a sufficient answer to this question.
I only found some rarely information: (Visit http://en.wikibooks.org/wiki/Data_Mining_Algorithms_In_R/Clustering/K-Means)
From this description Lloyd, Forgy, and Hartigan-Wong seem the same to me. Minimizing the within sum of squares or Minimizing the Euclidean Distance is the same.
MacQueen is different in the case that it updates the two involved clusters if a object is moved to another cluster if I'm right.
Nevertheless, I still do not see in which points these Algorithms are different.