An approach to constraint Inductive Logic Programming

A continuing problem with Inductive Logic Programming (ILP) has proved to be difficult to handle. Constraint Inductive Logic Programming (CILP) aims to solve this problem with ILP. We propose a new approach to CILP, and implement a prototype of CILP system called BPU-CILP. In our approach, methods from pattern recognition, such as Fisher's linear discriminant and prototype-based partitional clustering, are introduced to CILP. BPU-CILP can generate various forms of polynomial constraints in multiple dimensions, without additional background knowledge. As results, the CLP program covering all positive examples and consisting with all negative examples can be automatically derived.