There is a new version (2.0) of Bims (Bayesian inference of model structure) on the server.
Bims is an MCMC based tool for likelihood based learning of structured
statistical models. The library comes with likelihoods and priors for
classification trees and Bayesian networks.
This version implements the system as described in the following recent publication-
which is now the main reference for the software:
Distributional Logic Programming for Bayesian Knowledge Representation.
Nicos Angelopoulos and James Cussens.
International Journal of Approximate Reasoning (IJAR).
Volume 80, January 2017, pages 52-66.
Installation
?- pack_install(bims).
Testing
?- bims( [] ).
Documentation:
http://stoics.org.uk/~nicos/sware/bims/doc/html/bims.htmlWeb-page
http://stoics.org.uk/~nicos/sware/bims/Nicos Angelopoulos
---
http://stoics.org.uk/~nicos