Research Group
on Artificial Intelligence of the Hungarian Academy of Sciences
and of the University of Szeged
A substantial number of linear and nonlinear feature space transformation
methods have been proposed in recent years. Using the so-called "kernel-idea"
well-known linear techniques such as Principal Component Analysis(PCA),
Linear Discriminant Analysis(LDA) and Independent Component Analysis(ICA)
can be non-linearized in a general way. The aim of this paper here is
twofold. First, we describe this general non-linearization technique
for linear feature space transformation methods. Second, we derive formulas
for the ubiquitous PCA technique and its kernel version, first proposed
by Schölkopf et al., using this general schema and we examine how
this transformation affects the efficiency of several learning algorithms
applied to the phoneme classification task.