Induction from Answer Sets in Nonmonotonic Logic Programs
ACM Transactions on Computational Logic, vol.6(2), pages 203--231, 2005.
Inductive logic programming (ILP) realizes inductive machine
learning in computational logic.
However, the present ILP mostly handles classical clausal programs,
especially Horn logic programs, and has limited applications to
learning nonmonotonic logic programs.
This paper studies a method for realizing induction in nonmonotonic
We consider an extended logic program as a background theory, and
introduce techniques for inducing new rules using
answer sets of the program.
The produced new rules explain positive/negative examples in the context
of inductive logic programming.
The proposed methods extend the present ILP techniques to a syntactically
and semantically richer framework, and contribute to a theory of
Full Paper (Pdf 225K)