Bayesian Knowledge Discoverer

Bayesian Knowledge Discoverer (BKD) is a computer program able to learn Bayesian Belief Networks from (possibly incomplete) databases. BKD is based on a new estimation method called Bound and Collapse and it has been developed within the Bayesian Knowledge Discovery project. BKD has been distributed in over 3,000 copies around the world.

BKD has now a commercial successor, Bayesware Discoverer, distributed by Bayesware Limited. A free copy of the program for non-commercial use is available from the company website.

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

Estimation: Estimation of conditional probability distributions.
Model Selection: Bayesian learning of the graphical structure.
Propagation: Provide a goal oriented propagation algorithm.
Discretization: Unsupervised discretization continuos variables.
Missing Data: Handle missing data using Bound and Collapse.
Automated Definition: Automated variable definition from data.
BNIF Interface: Import/export for BN Interchange Format.
User Interface: Interactive GUI and search process animation.
Documentation: Movie-based on-line help.

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

Platforms: BKD Version 1.0 (Beta) for MS Windows 9x/NT (8.9 MB).

Availability: The commercial successor of BKD, Bayesware Discoverer, is now distributed by Bayesware Limited. A free copy of the program for non-commercial use is available from the company website.