SIGMA : Jurnal Sains Dan Teknologi, Vol. 11, No. 2, Juli 2008, ABSTRAK : K-mean method is a clustering method in which grouping techniques are based only on distance measure among observed objects, without considering statistical aspects. Model-based clustering is a method that uses statistical aspects, as its theoretical basis i.e.probability maximun criterion. This mode has several variations with a variety of geometrical characteristics obtained by mean Gauss component. Dat partition is conducted by utilizing expectation-maximization (EM) algorithm. ...