Inductive Logic Programming (ILP) is known as a promising technique used by Explainable AI; however, it is also known as inefficient. To address this problem, we propose a new approach that takes advantage of the neural network trained over examples. The use of the neural network makes the hypothesis search in rules and knowledge faster. It allows inductive reasoning to take advantage of the generated hypothesis as background knowledge. To demonstrate the effectiveness of our approach, we have implemented an ILP system and evaluated it. The experimental results show that our approach remarkably decreased its execution time for bigdata while maintaining accuracy.