FISHERC Trainable classifier: Fisher's Least Square Linear Discriminant
W = FISHERC(A)
DescriptionFinds the linear discriminant function between the classes in the dataset A by minimizing the errors in the least square sense. This is a multiclass implementation using the oneagainstall strategy. It results in a set of linear base classifiers, one for every class. The final result may be improved significantly by using a nonlinear trained combiner, e.g. by calling W = A*(FISHERC*QDC([],[],1e6);
For high dimensional datasets or small sample size situations, the PseudoFisher procedure is used, which is based on a pseudoinverse. This classifier, like all other nondensity based classifiers, does not use the prior probabilities stored in the dataset A. Consequently, it is just for twoclass problems and equal class prior probabilities equivalent to LDC, which assumes normal densities with equal covariance matrices. Note that A*(KLMS([],N)*NMC) performs a very similar operation, but uses the prior probabilities to estimate the mean class covariance matrix used in the prewhitening operation performed by KLMS. The reduced dimensionality N controls some regularisation. Reference(s)1. R.O. Duda, P.E. Hart, and D.G. Stork, Pattern classification, 2nd ed. John Wiley and Sons, New York, 2001. See alsomappings, datasets, testc, ldc, nmc, fisherm,
