Inductive Logic Programming [42.1] is the research area formed at the intersection of logic programming and machine learning. Rough set theory [42.2], [42.3] defines an indiscernibility relation, where certain subsets of examples cannot be distinguished. The gRS-ILP model [42.4] introduces a rough setting in Inductive Logic Programming and describes the situation where the background knowledge, declarative bias and evidence are such that it is not possible to induce any logic program from them that is able to distinguish between certain positive and negative examples. Any induced logic program will either cover both the positive and the negative examples in the group, or not cover the group at all, with both the positive and the negative examples in this group being left out.