Maximizing margins.
Soft margins.
Kernels.
Kernels everywhere.
Many researchers have first approached SVMs with Chris Burges' tutorial.
Section 1 to 4 of the Bottou and Lin chapter propose a self-contained
mathematical treatment of duality in SVMs.
The most used softwares for SVMs
are LibSVM
and SVMLight.
Also the SVQP2 solver is
a small and efficient C++ library for solving the SVM dual optimization problem.
Note that one should use very different algorithms for training SVMs with kernels
and SVMs with linear parametrization.