Billow depends on the Pepl implementation of the FAM ([1]) algorithn, version 0:0:4 or later.
This can be obtained from here. (Install Pepl and Billow in the same directory.)

Current version is available from this location: billow-0_0_3-2003ab28.tgz Gunzip, untar, cd to the newly created directory.
Look around and then read doc/user_guide.ps. An example of using the software is described in [2] and [3].

Acknowledgments
This work was supported partly by the ESPRIT IST project ``Application of Probabilistic Inductive Logic Programming (APRIL)'', (Grant reference IST-2001-33053) and the BBSRC/EPSRC Bio-informatics and E-Science Programme, ``Studying Biochemical networks using probabilistic knowledge discovery'' (Grant reference 28/BEP17011).

References:

Failure Adjusted Maximasation explained :

[1] James Cussens. Parameter estimation in stochastic logic programs. Machine Learning, 44(3):245-271, 2001

Example application for the software :

[2] Nicos Angelopoulos and Stephen Muggleton. Machine learning metabolic pathway descriptions using a probabilistic relational representation. In Machine Intelligence 19, September 18-20 2002.

[3] Nicos Angelopouloos and Stephen Muggleton. SLPs for probabilistic pathways: Modelling and parameter estimation. Technical report, Department of Computing, Imperial College, London, 2002.