Inductive Logic Programming(ILP) is a symbolic rule learning method that introduces the nesting of functions and logical expressions in first-order rule learning and uses first-order logic as the expression language.

ILP enables machine learning systems to have more powerful expressive capabilities. At the same time, it can be seen as an application of machine learning, mainly used to solve the induction of logic programs based on background knowledge. The relevant rules can be directly used by logic programming languages such as PROLOG.

The ILP related design architecture is as follows:

Positive examples + negative examples + background knowledge ⇒ hypothesis

The model learned by ILP is based on the symbolic rules of first-order logic rather than an incomprehensible black box model. The learned model can be based on the relationship between individuals rather than just predicting the individual's label.

References

【1】Inductive Logic Programming (personal blog)

【2】A Survey of Inductive Logic Programming