Bayes Classifier Induction

Download

Attention: In order to compile these programs the table package must also be retrieved. The table package also contains some auxiliary programs for preprocessing the data files.

Description

Programs to induce a full or a naive Bayes classifier and to classify new data with an induced full or naive Bayes classifier.

More details about these programs (included in the source package).

A program (bcview) to visualize the classifiers induced with these programs can be found here.

A graphical user interface to these command line programs (BayesGUI, written in Java) can be found here.

Another graphical user interface to these programs, which is based on Gnome 2 and was developed by togaware can be found here. (However, I cannot guarantee that this GUI works with the latest version of the command line programs made available here.)

A review of the theory underlying Bayes classifiers can be found in:

• Graphical Models - Methods for Data Analysis and Mining
  Christian Borgelt and Rudolf Kruse.
  J. Wiley and Sons, Chichester, United Kingdom 2002, ISBN 0-470-84337-3
• A Naive Bayes Classifier Plug-In for DataEngine.
  Christian Borgelt.
  Proc. 3rd Data Analysis Symposium, 87-90.
  MIT GmbH, Aachen, Germany 1999
  nbplugin.pdf (144 kb) nbplugin.ps.gz (114 kb) (4 pages)
• Probabilistic Networks and Fuzzy Clustering as Generalizations of Naive Bayes Classifiers.
  Christian Borgelt, Heiko Timm, and Rudolf Kruse.
  Computational Intelligence in Theory and Practice (Advances in Soft Computing), 121-138
  Bernd Reusch and Karl-Heinz Temme, eds.
  Physica-Verlag, Heidelberg, Germany 2001
  dofd_98.pdf (334 kb) dofd_98.ps.gz (219 kb) (4 pages)