[LABELS,W_EM] = EMCLUST (A,W_CLUST,K,LABTYPE)
The untrained classifier mapping W_CLUST is used to update an initially labeled dataset A by iterating the following two steps
If K is given, a random initialisation for K clusters is made and labels of A are neglected. If K is omitted the given labeling is used as initailisation.
An early stopping of the EM algorithm is controlled by PRTIME.
LABTYPE determines the type of labeling: 'crisp' or 'soft'. Default: label type of A. It is assumed W_CLUST can handle the LABTYPE requested. Only in case LABTYPE is 'soft' the traditional EM algorithm is followed. In case LABTYPE is 'crisp' EMCLUST follows a generalised k-means algorithm.