Induction of logical relations based on specific generalization of strings

Learning logical relations from examples expressed as first order facts has been studied extensively by the inductive logic programming research. Learning with positive-only data may cause over generalization of examples leading to inconsistent resulting hypotheses. A learning heuristic inferring specific generalization of strings based on unique match sequences is shown to be capable of learning predicates with string arguments. This paper describes an inductive learner based on the idea of specific generalization of strings, and the given clauses are generalized by considering the background knowledge.