Hendrik Blockeel

Address : K.U.Leuven, Dept. of Computer Science
          Celestijnenlaan 200 A
          B-3001 Heverlee, Belgium

email:    Hendrik.Blockeel@cs.kuleuven.ac.be

Title: From machine learning to Inductive Logic Programming: ILP made easy

Type: foundational

Section: Logic and Computation

Description:

This course is about the use of inductive logic programming for
machine learning and data mining.  It is assumed that the audience is
familiar with computational logic, but not necessarily with inductive
reasoning.  A short introduction on the latter will be given: what is
it, what are its applications ?  We focus on applications in machine
learning and data mining.  We briefly discuss the basics of ILP, and
discuss issues such as data representation, definition of the
hypothesis space, algorithms.  Next an overview of applications of ILP
in machine learning and data mining is given, more specifically how
ILP can be integrated with existing methods for association rule
discovery, reinforcement learning etc. and what advantages it offers
in these contexts.  Finally we discuss the relevance of ILP outside
computational logic, and some efforts that have been done to make ILP
technology usable for people not familiar with computational logic.