Product combining classifier
W = PRODC(V)
It defines the product combiner on a set of classifiers, e.g. V=[V1,V2,V3] trained on the same classes, by selecting the class which yields the highest value of the product of the classifier outputs. This might also be used as A*[V1,V2,V3]*PRODC in which A is a dataset to be classified.
If it is desired to operate on posterior probabilities, then the input classifiers should be extended like V = V*CLASSC.
The base classifiers may be combined in a stacked way (operating in the same feature space) by V = [V1,V2,V3,...] or in a parallel way (operating in different feature spaces) by V = [V1;V2;V3;...].
mappings, datasets, votec, maxc, minc, medianc, meanc, averagec, stacked, parallel,