SMO: "Implements John C. Platt's sequential minimal optimization algorithm for training a support vector classifier using polynomial or RBF kernels. This implementation globally replaces all missing values and transforms nominal attributes into binary ones. It also normalizes all attributes by default. (Note that the coefficients in the output are based on the normalized/standardized data, not the original data.) Multi-class problems are solved using pairwise classification. To obtain proper probability estimates, use the option that fits logistic regression models to the outputs of the support vector machine. In the multi-class case the predicted probabilities will be coupled using Hastie and Tibshirani's pairwise coupling method. Note: for improved speed standardization should be turned off when operating on SparseInstances."
'via Blog this'
