[mercury-users] inductive mercury programming

Barney Fisher barney.fisher at googlemail.com
Thu May 17 19:31:59 AEST 2007

Hello Mercury users,

James Cussens and I have implemented an inductive logic programming
system in Mercury and we thought this might be of interest to the
Mercury team.  A paper summarising the system with results was
presented at ILP 06, the abstract is as follows:

We investigate using the Mercury language to implement
and design ILP algorithms, presenting our own ILP system IMP. Mer-
cury provides faster execution than Prolog. Since Mercury is a purely
declarative language, run-time assertion of induced clauses is prohibited.
Instead IMP uses a problem-speciļ¬c interpreter of ground representa-
tions of induced clauses. The interpreter is used both for cover testing
and bottom clause generation. The Mercury source for this interpreter
is generated automatically from the user's background knowledge using
Moose, a Mercury parser generator. Our results include some encourag-
ing results on IMP's cover testing speed, but overall IMP is still generally
a little slower than ALEPH.

A full version of the paper can be found at

If you're not familiar with the term Inductive Logic Programming (ILP)
it can be summarised as a type of learning which takes as input
examples and background knowledge as a logic program and outputs
another logic program, which w.r.t. the background knowledge explains
the examples.  The learnt program should be a generalisation of the

We invite feedback of our work, in particular the implementation
details which can be found in sections 4 and 5 of the paper.
Hopefully you can tell us 'whether we're doing it right'.


mercury-users mailing list
Post messages to:       mercury-users at csse.unimelb.edu.au
Administrative Queries: owner-mercury-users at csse.unimelb.edu.au
Subscriptions:          mercury-users-request at csse.unimelb.edu.au

More information about the users mailing list