Dd_tools

Current version is 1.9.1 (May 2012).

The data description toolbox wants to provide tools, classifiers and evaluation functions for the research of one-class classification (or data description). 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 precision-recall curve and mean precision, and many classifiers.

The toolbox is an extension of the PRTools toolbox, in which Matlab objects for mapping and dataset are defined. The data description toolbox uses these objects and their methods, but extends (and sometimes restricts) it to one-class classification. This means, that before you can use it to its full potential, you need to know a bit about PRTools. When you are completely new to pattern recognition, Matlab or PRTools, please familiarize yourself a bit with them first.

Contents where you can see all the function descriptions (which acts more or less as a help).

Online dd_tools manual (in HTML format).

Dd_tools manual (in PDF format). This manual is also included in the ZIP-file below.

Citation

If you enjoyed using it and used it for making a nice publication, then you are kindly requested to refer to the dd_tools:

@misc{Ddtools2012,
   Author = {Tax, D.M.J.},
   Month = {May},
   Note = {version 1.9.1},
   Title = {DDtools, the Data Description Toolbox for Matlab},
   Year = {2012}}

Download

In this version dd_tools 1.9.1 some small improvements on the incremental support vector approaches are made. Furthermore, the multic.m and ocmcc.m are extended to also allow for example outlier objects during training. And some small tunings, and bugfixes are made.

For this version of dd_tools you need PRTools 4.0.22 or higher PLEASE update to the new PRTools 4.0

Finally, you can download the important things, software and manual:

• Download toolbox version 1.9.1.
• Actually, I realize that things change quite often, and I'm not always very fast in updating this webpage. But I made a pdf document explaining the very basics of the toolbox. You can get the manual here.

Have fun!