Inductive Classification Logic

Inductive Classification Logic (ICL in short) is a ILP learning system that learns first order logic formulae from examples which belong to two or more classes. The learned theory can be used to classify unseen examples. Examples are viewed as (Herbrand) interpretations. These are assumed be specified completely (we also say that we learn from closed examples). So ICL performs discriminating induction from closed examples.

ICL is being developed by Wim Van Laer and Luc De Raedt. Both work in the Machine Learning Group at the departement of Computer Science of the Katholieke Universiteit Leuven. They are supported by the FWO (Fund for Scientific Research - Flanders, Belgium).

These Web pages contain some information on the system ICL. If you miss something, don't hesitate and contact Wim Van Laer. All comments are welcome.

• Contact information
• Obtaining ICL
• Installing the system
• User Guide
  • General Information
  • Starting up
  • The Input Files
  • Interaction with the system
  • Overview of the commands
  • The Output Files
  • Special features
  • Some small examples
  • Tips and Tricks
• Papers
• Experiments (only locally available)

Copyright 1998, Katholieke Universiteit Leuven, dept. Computerwetenschappen Information provider: KULeuven dept. Computerwetenschappen Comments for the authors: Wim Van Laer Page design: Wim Van Laer URL: http://www.cs.kuleuven.ac.be/main.html