RISO Jellyfish

RISO provides an implementation of distributed, heterogeneous belief networks.

  Add it to your Download Basket!

 Add it to your Watch List!


Rate it!
send us
an update
GPL (GNU General Public License) 
3.7/5 15
Robert Dodier
ROOT \ Science and Engineering \ Mathematics
RISO provides an implementation of distributed, heterogeneous belief networks.

RISO is an implementation of heterogeneous, distributed belief networks in Java. A belief network is a probability model defined on an acyclic directed graph; distributed means nodes can be on different hosts, and heterogeneous means allowing different conditional distributions.

The calculations involved are multidimensional integrations; exact results are known for a catalog of special cases. If a partial result cannot be calculated as a special case from the catalog, RISO computes an approximate result by numerical integration.

Partial results are passed from one node in the graph to another as messages; if nodes live on different hosts, the belief network is said to be distributed. Messages are passed via RMI. Many example belief networks and lengthy documents are included in the RISO release bundle.

What's New in This Release:

The "Jellyfish" release of RISO contains minor bug fixes and some functional enhancements.
There is a Python module riso_binding.py which can import RISO belief networks into Jython (Python implemented in Java).
Jython is a more flexible and elegant user interface than riso/apps/RemoteQuery.java.
The compile.sh script uses JavaDeps to generate a makefile (javac wasn't capable of resolving all the cross dependencies). You'll need JavaDeps only if you want to do a clean recompile; recompiling one file at a time with javac (given all the other class files present) has always succeeded so far.
There are now Poisson, exponential, and binomial distributions.
There is initial support for the notion of considering a belief network as a multidimensional distribution. See riso/distributions/Factorized.java.
There is an initial version of a Gibbs sampler algorithm. See riso/apps/GibbsSampler.java.

Last updated on April 18th, 2007

#heterogeneous Bayesian #belief networks #acyclic directed graph #RISO #heterogeneous #Bayesian #acyclic

Add your review!