A New Algorithm for Computing Least Generalization of a Set of Atoms
Hien D. Nguyen and Chiaki Sakama
in: Proceedings of the
29th International Conference on Inductive Logic Programming (ILP 2019),
Lecture Notes in Artificial Intelligence 11770, pages 81-97, Springer-Verlag, 2019.
This paper provides a new algorithm of computing a least generalization of a set of atoms.
Our algorithm is based on the notion of anti-combination that is the inverse substitution of
a combined substitution.
In contrast to an anti-unification algorithm that computes a least generalization of two atoms,
anti-combination can compute a least generalization of (more than two) atoms in parallel.
We evaluate the proposed algorithm using randomly generated data and show that anti-combination
outperforms the iterative application of an anti-unification algorithm in general.
Full Paper (PDF 1220K)