Introduction dd_tools

In dd_tools it is possible to define special one-class datasets and one-class classifiers. Furthermore, the toolbox provides methods for generating artificial outliers, estimating the different errors the classifiers make (false positive and false negative errors), estimating the ROC curve, the AUC (Area under the ROC curve) error, the AUC over a limited integration domain and many classifiers.

This is reflected in the setup of this manual. Each of the ingredients are discussed in a separate section:

1. One-class datasets,
2. One-class classifiers,
3. One-class error evaluation or model selection.

Before you can use the toolbox, you have to use Prtools and you have to put your data into a special dataset-format. Let us first start with the data.