5. Nonlinear classification problems

5.1. Extension of Adaline

Todo

Under construction. See for example [raschka_python_2015].

• Example for non-linear separable data that can be mapped to a higher dimensional feature space where the data becomes linear separable.

5.2. Kernel trick for Support Vector Machines

Todo

Under construction. See for example [vapnik_statistical_1998], [mohri_foundations_2012].

• Properties of kernels and relation between them and a mapping to a higher dimensional feature space.
• SVMs with Kernels and there advantages for non-linear problems.
• Normalization of kernels.