Nearest Mean Classifier
W = NMC(A)
Computation of the nearest mean classifier between the classes in the dataset A. The use of soft labels is supported. Prior probabilities are not used.
The difference with NMSC is that NMSC is based on an assumption of normal distributions and thereby automatically scales the features and is sensitive to class priors. NMC is a plain nearest mean classifier for which the assigned classes are are sensitive to feature scaling and unsensitive to class priors.
The estimated class confidences by B*NMC(A), however, are based on assumed spherical gaussian distributions of the same size.
As NMC is a simple linear classifier, a non-linear combiner might give a significant performance improvement in multi-dimensional problems, e.g. by W = A*(NMC*QDC(,,1e-6)).