Introducing Inductive Methods in Knowledge Acquisition by Divide-and-Conquer

Contents

In this paper we present the outline of a method that combines a divide-and-conquer approach, that is common in knowledge acquisition and software engineering, with the use of inductive techniques. The idea is to guide decomposition of a knowledge acquisition problem into subpr oblems by the expected costs and benefits of applying elicitation or induction t o acquire the knowledge for part of the target knowledge. The method is illustra ted with a "rational reconstruction" of a knowledge acquisition process that involved inductive techniques.