RBSTATSSVC

Trainable automatic radial basis Support Vector Classifier

    W = RBSTATSSVC(A)
    W = A*RBSTATSSVC

Input
A	Dataset

Output
W	Mapping: Radial Basis Support Vector Classifier

Description

This routine optimizes a radial basis kernel for STATSSVC. The kernel is  based on PROXM. As STATSSVC is basically a two-class classifier, solving  multi-class problems in a one-against-rest fashion, different kernels may  be needed for every base classifier. The result W is thereby a stacked  combiner.

If any class in A has less than 20 objects, the kernel is not optimised  but estimated by PKSTATSSVM, using the Parzen kernel.

See also

mappings, datasets, proxm, svc, statssvc, pkstatssvc, rblibsvc, rbsvc,