Inductive Logic Programming and Multistrategy Learning 1
Claude Sammut · 1996
This paper discusses the role that inductive logic programming can play in building flexible multistrategy learning systems. We demonstrate that a variety of learning algorithms can be embodied in the background knowledge of an ILP system. ILP is an effective framework for multistrategy learning because it has a mechanism for learning new concept descriptions which can refer to knowledge provided by the user or learned from some other task. The Horn clause representation is central to this mechanism since it can be used to specify background knowledge in such a way that the learner can apply its knowledge to new information.