(Apologies in advance if you have received this from other newsgroups
in intelligent systems or applied mathematics and statistics.)

The Kansas State University Laboratory for Knowledge Discovery in
Databases is pleased to announce a new release of Bayesian Network
Tools in Java (BNJ) v.2:
http://bndev.sourceforge.net

WHAT'S NEW:
v2.03a is the penultimate alpha release of v2 and now has full GUI
features.
A tutorial is available on the SourceForge and the Yahoo! Group
(bndev).
The first edition of the BNJ 2 user manual is also available from the
BNJ distribution page.

BNJ is an open-source suite of Java tools for research and
applications development in probabilistic reasoning, and currently
includes the following modules:

- Graphical editor for networks and conditional probability tables
(CPTs)
- Exact inference (junction tree, variable elimination)
- Sampling-based inference (logic sampling, importance sampling)
- K2 algorithm for structure learning
- Experimental packages for structure learning
(stochastic score-based structure search, variable ordering)
- Format interconversion suite

The latest stable release of BNJ is always available on Sourceforge
and can be downloaded with or without Javadoc.

BNJ is compatible with the XML Bayesian Network interchange format and
can convert among many network formats, including commercial (Hugin,
Netica, Microsoft BN) and academic ones (U. Pittsburgh Genie, U. Sao
Paulo Bayesian Interchange Format, Hebrew University LibB).

For more information, please see the BNJ distribution page.
BNJ is distributed under the GNU General Public License.

As always, we would like to invite anyone interested in developing
intelligent systems with graphical models of probability to join our
Bayesian Network tools development Yahoo! Group:
http://groups.yahoo.com/group/bndev

-William Hsu