A learning simulator for testing CAI systems

Abstract

The authors present a learning simulator consisting of an interface, an inference engine and an inductive logic programming (ILP) system. Possible usage of the simulator includes checking the behavior of CAI systems, to be adopted as a novice agent in CAI systems, and for a teacher to check contents for study by observing the response of the simulator. The learning simulator learns interactively. First, a teacher (or a CAI system) gives background knowledge as basic rules and examples. Next, a teacher asks the simulator some question. Using the background knowledge, rules generated by ILP and examples stored in the memory, the simulator answers the question. Then, the simulator stores the examples and updates rules for the next question, after the teacher tells the correct answer. We implement the simulator using a Prolog interpreter and ILP system FOIL. We show learning results obtained through computer simulations.