PKLIBSVC

Automatic radial basis LIBSVM, using NULIBSVC and the Parzen kernel

    [W,KERNEL] = PKLIBSVC(A,ALF)
    [W,KERNEL] = A*PKLIBSVC([],ALF)
    [W,KERNEL] = A*PKLIBSVC(ALF)

Input
A	Dataset
ALF	Parameter, default 1

Output
W	Mapping: Radial Basis Support Vector Classifier
KERNEL	Untrained mapping, representing the optimised kernel

Description

This routine provides a radial basis support vector classifier based on  LIBSVC. It estimates the kernel width SIGMA by the the value found by  PARZENC. The kernel width used is ALF*3*SQRT(2)*SIGMA. This is much  faster than the gridsearch used by RBLIBSVC and may performs equally well.

Note that LIBSVC is a multi-class classifier and thereby uses a single  kernel.

See also

datasets, mappings, libsvc, nulibsvc, rblibsvc, parzenc, svc,