Oracle Support Vector Machine (SVM)

Support Vector Machine (SVM) is a classification and regression algorithm that uses machine learning theory to maximize predictive accuracy without overfitting the data. SVM uses an optional nonlinear transformation of the training data, followed by the search for regression equations in the transformed data to separate the classes (for categorical targets) or fit the target (for continuous targets). Oracle’s implementation of SVM allows models to be built by using one of two available kernels, linear or Gaussian. The linear kernel omits the nonlinear transformation altogether so that the resulting model is essentially a regression model.

For more information, see the Oracle Data Mining Application Developer’s Guide and Oracle Data Mining Concepts.