Bayesian network structure learning using an hybrid approach

Here below is a free software written in R by Maxime Gasse (Ph.D. student) that implements an efficient and scalable hybrid algorithm for Bayesian network structure learning, called Hybrid HPC (H2PC). It first reconstructs the skeleton of a Bayesian network and then performs a Bayesian-scoring greedy hill-climbing search to orient the edges. The software implements several procedures discussed in our recent ECML-PKDD 2012 article. It requires several routines implemented in the bnlearn package in R developped by Marco Scutari. The source code of H2PC as well as all data sets used for the empirical tests are freely available. The code is distributed in the hope that it will be useful to researchers and bayesian netwok practitionners, but without any warranty; without even the implied warranty of merchantability or fitness for a particular purpose. This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation.

  Download H2PC latest version (July 1st 2012)

  Download the data sets used in the ECML paper

About Me

Alexandre Aussem

Alexandre Aussem

Professor in Computer Science


Address: Batiment Nautibus, 43 bd du 11 novembre 1918, 69622 Villeurbanne Cedex.
Phone: +33 4 26 23 44 66
Fax: +33 4 72 43 15 37
Email: aaussem at univ-lyon1 dot fr

To visit

Université Claude Bernard Lyon 1