Efficient Learning of Logic Programs with Non-determinant, Non-discriminating Literals

We propose a new heuristic-based approach to Horn-clause logic program learning. The heuristic can be viewed as an improvement in the information-based estimate employed in FOIL, and captures the important goal-directed usefulness of a literal which is overlooked in it. Our system, CHAM, learns a class of complex programs not learned by previous systems, i.e., non-determinate programs out of the learning space of GOLEM, and programs with non-discriminating literals which pose difficulties for FOIL. While being able to learn the larger class of programs, CHAM is shown to preserve efficiency in various test problems.