The basic C_SVC SVM type. The default, and a good starting point
The NU_SVC type uses a different, more flexible, error weighting
One class SVM type. Train just on a single class, using outliers as negative examples
A SVM type for regression (predicting a value rather than just a class)
A NU style SVM regression type
A very simple kernel, can work well on large document classification problems
A polynomial kernel
The common Gaussian RBD kernel. Handles non-linear problems well and is a good default for classification
A kernel based on the sigmoid function. Using this makes the SVM very similar to a two layer sigmoid based neural network
A precomputed kernel - currently unsupported.
The options key for the SVM type
The options key for the kernel type
Training parameter, boolean, for whether to use the shrinking heuristics
Training parmater, boolean, for whether to use probability estimates
Algorithm parameter for Poly, RBF and Sigmoid kernel types.
The option key for the nu parameter, only used in the NU_ SVM types
The option key for the Epsilon parameter, used in epsilon regression
Training parameter used by Episilon SVR regression
Algorithm parameter for poly and sigmoid kernels
The option for the cost parameter that controls tradeoff between errors and generality
Memory cache size, in MB