Combining of linear classifiers by averaging coefficients
W = AVERAGEC(V)
Let V = [V1,V2,V3, ... ] is a set of affine classifiers trained on the same classes, then W is the average combiner: it averages the coefficients of the base classifiers, resulting in a new affine classifier. This might also be used as A*[V1,V2,V3]*AVERAGEC, in which A is a dataset to be classified.
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, prodc, averagec, stacked, parallel,