BNet.News
In This Issue: September/October 2006 
•   BNet.Builder 1.4 Release
•   BNet.EngineKit User Tip: Adding a Node
•   Downloading BNet.Builder 1.4
•   Charles River Analytics' Presentations at UAI 2006
•   BNet.Builder User Tip: Expand/Collapse Beliefs and Evidence
•   Publicize Your Belief Network Success Story
BNet.Builder 1.4 Release
CAMBRIDGE, MA - Charles River Analytics recently released version 1.4 of their Bayesian belief network modeling tool, BNet.Builder. BNet.Builder is a desktop application for rapidly creating Belief Networks, entering information, and getting results. BNet.Builder allows users to create models quickly and easily.

Zach Cox, software engineer and chief developer of BNet.Builder said, "BNet.Builder 1.4 includes a new state-of-the-art inference engine (based on Java bytecode generation) and several important bug fixes."

BNet.Builder is part of Charles River Analytics' family of Bayesian Network products, including BNet.EngineKit.

BNet.Builder is available directly from Charles River Analytics. For additional information or a free trial version:

* Contact Zach Cox at (617) 491-3474 x523 or e-mail bnet-sales@cra.com

* Or visit: http://www.cra.com/bnet.builder

BNet.Builder
Downloading BNet.Builder 1.4
If you've downloaded an earlier trial version of BNet.Builder and want to upgrade to the trial version of BNet.Builder 1.4, follow these steps:

* Uninstall the old version (run the uninstaller from the Start Menu/BNet.Builder/Uninstall BNet.Builder).
* Install the new version (available at http://www.cra.com/bnet.builder).

For the full commercial version of BNet.Builder, visit http://www.cra.com/bnet.builder

BNet.Builder
BNet.Builder User Tip: Expand/Collapse Beliefs and Evidence
Each node displayed in BNet.Builder has buttons to expand or collapse both beliefs and evidence. However, if you want to expand or collapse the beliefs or evidence on many nodes at once, clicking the button on each node can become very tedious. BNet.Builder 1.3 solved this problem by introducing four new commands that expand or collapse the beliefs or evidence on all selected nodes.

To use these new commands, select multiple nodes and choose Show beliefs, Show evidence, Hide beliefs, or Hide Evidence in the View menu. Each of these commands also has a toolbar button (look for the same icons used in the View menu).

To learn more about BNet.Builder 1.4, click here.

BNet.EngineKit User Tip: Adding a Node
The BayesianNetwork class in BNet.EngineKit provides two methods for adding a node. The first method adds a node with a specified name that has the default states true and false. The following code shows how to add a new node named Alarm with states true and false:

BayesianNetwork network = ...;
DiscreteNode node = network.addNode("Alarm");

To add a node with states other than true and false, the BayesianNetwork class provides a second form of the addNode method. The following code shows how to add a node named Temperature with states high, medium, and low:

BayesianNetwork network = ...;
List states = new ArrayList();
states.add("high");
states.add("medium");
states.add("low");
DiscreteNode node = network.addNode("Temperature", states);

You can also use the Arrays utility class that comes with the JDK to simplify this a bit:

BayesianNetwork network = ...;
DiscreteNode node = network.addNode("Temperature", Arrays.asList(new String[] {"high", "medium", "low"}));

You can pass any List object into this second form of the addNode method. And remember that you can fully modify the list of states returned by the DiscreteNode.getStates() method.

To learn more about Engine.Kit, click here.

Bnet.EngineKit
Charles River Analytics' Presentations at UAI 2006
CAMBRIDGE, MA – Charles River Analytics presented research on Bayesian belief networks at the Fourth Bayesian Modeling Applications Workshop during the 22nd Conference on Uncertainty in Artificial Intelligence (UAI 2006).The conference took place from July 13-16, 2006, at the Massachusetts Institute of Technology in Cambridge, Massachusetts. The theme of this year’s Bayesian Modeling Applications Workshop was “Bayesian Models meet Cognition.”

During the session on Modeling Human Factors, Charles River Analytics presented two workshops. Sean Guarino presented “Modeling Human Reasoning about Meta-Information.” David Koelle presented “Applications of Bayesian Belief Networks in Social Network Analysis.”

To view papers from the workshop, click here.

Publicize Your Belief Network Success Story
Want to show off your applications of belief networks to others using them in their research? Send us your success stories. We would like to include at least one in each edition of BNet.News.

To send us your story, please contact lcordeiro@cra.com or zcox@cra.com.

BNet News is published by Charles River Analytics, the company that brings you BNet.Builder and BNet.EngineKit.  View our Privacy Policy