Improving the efficiency of ILP systems

Abstract

Inductive Logic Programming (ILP) is a promising technol- ogy for knowledge extraction applications. ILP has produced intelligible solutions for a wide variety of domains where it has been applied. The ILP lack of eciency is, however, a major impediment for its scalability to applications requiring large amounts of data. In this paper we pro- pose a set of techniques that improve ILP systems eciency and make then more likely to scale up to applications of knowledge extraction from large datasets. We propose and evaluate the lazy evaluation of examples, to improve the eciency of ILP systems. Lazy evaluation is essentially a way to avoid or postpone the evaluation of the generated hypotheses (coverage tests). The techniques were evaluated using the IndLog system on ILP datasets referenced in the literature. The proposals lead to substantial eficiency improvements and are generally applicable to any ILP system.