RoC: The Robust Bayesian Classifier.

The Robust Bayesian Classifier (RoC) is a computer program able to perform supervised Bayesian classification from incomplete databases, with no assumption about the pattern of missing data. RoC is based on a new estimation method called Robust Bayesian Estimator and it was  developed within the Bayesian Knowledge Discovery project.

Capabilities: Take a tour to explore the capabilities of RoC Version 1.0.

Training: Estimate  conditional probability distributions.
Testing:  Predict class label of cases in a database.
Discretization: Continuos variables are discretized.
Missing Data: Handle incomplete cases.
Cross validation: A cross validation utility for evaluation.
Automatic Definition: Attributes are automatically defined from data.
User Interface: Easy-to-use wizard interface.
Documentation:  On screen context sensitive help.
Documentation: An Introduction to the Robust Bayesian Classifier.

Paper Work: Read  the Release Notes and the Conditions of Use of RoC.

Platforms: RoC Version 1.0 (Beta) for Microsoft Windows 9x/NT (4.9 MB)