Abstract
A logic-based approach to Machine Learning, called Inductive Logic Programming (ILP), is outlined, and several applications of ILP are reviewed. These benefit from the ability of ILP to flexibly use background knowledge in the learning process. The results from a variety of experimental applications of ILP are presented together with a discussion of the relative advantages of ILP and other Machine Learning approaches. Experience shows that ILP is a powerful tool for ``knowledge-intensive'' data mining.