Constraint Inductive Logic Programming

This paper is concerned with learning from positive and negative examples expressed in rst-order logic with numerical constants. The presented approach is based on the cooperation of Inductive Logic Programming (ILP) and Constraint Logic Programming (CLP), and proceeds as follows: A discriminant induction problem is shown to be equivalent to a Constraint Satisfaction Problem (CSP): all constrained clauses covering positive examples and rejecting negative examples can be trivially derived from the solutions of this CSP. Solving this CSP then allows to build the G set of solutions in terms of Version Spaces; this resolution can be delegated to a constraint solver. This CSP provides a tractable computational characterization of G, which is suucient to classify further examples and ooers simple counting-based heuristics to resist noisy data. In this hybrid ILP-CLP approach, CLP performs most of the search involved in inductive learning; the advantage is to beneet from the number-handling capabilities of CLP, without requirement for additional background theory.