On avoiding redundancy in inductive logic programming

ILP systems induce rst-order clausal theories performing asearch through very large hypotheses spaces containing redundant hypotheses.The generation of redundant hypotheses may prevent the systemsfrom nding good models and increases the time to induce them.In this paper we propose a classication of hypotheses redundancy andshow how expert knowledge can be provided to an ILP system to avoidit. Experimental results show that the number of hypotheses generatedand execution time are reduced when expert knowledge is used to avoidredundancy