An Examination of the Third Stage in the Analogy Process: Verification-based Analogical Learning

Many studies of analogy in Artificial Intelligence have focused on analogy as a heuristic mechanism to guide search and simplify problem solving or as a basis for forming generalizations. This paper examines analogical learning, where analogy is used to conjecture new knowledge about some domain. A theory of Verification-Based Analogical Learning is presented which addresses the tenuous nature of analogically inferred concepts and describes procedures that can be used to increase confidence in the inferred knowledge. The theory describes how analogy may be used to discover and refine scientific models of the physical world. Examples are taken from an implemented system, which discovers qualitative models of processes such as liquid flow and heat flow.