Stephen W. Smoliar

Explanation: A Source of Guidance for Knowledge Representation

In constructing an expert system, there are usually several ways to represent a given piece of knowledge regardless of the knowledge representation formalism used. Initially, all of them may appear to be equivalent, but as the system evolves, it often becomes apparent that some are better than others, leading to the need to revise representations. Such revisions can be very time-consuming and prone to error. In this paper, we argue that the additional constraints imposed by the addition of an explanation facility can guide the creation of a knowledge base in a manner that reduces the need to subsequently re-structure the knowledge base as the systems functionality increases. We describe criteria that may be applied after the knowledge base is constructed to reveal potential weaknesses as well as those that may be employed during knowledge base construction. Finally, we briefly describe an expert system shell we have constructed that embodies these guiding principles, the Explainable Expert Systems framework.

Explaining the Link Between Causal Reasoning and Expert Behavior

There is a paradox in causal reasoning. It can be a powerful tool when making a difficult diagnosis1 and is frequently used when explaining why a particular diagnosis is correct. Diagnosticians in a variety of domains, however, do not seem to use it much when performing routine diagnoses.2 Why not?

Neural darwinism: The theory of neuronal group selection: Gerald M. Edelman, (Basic Books; New York, 1987); xxii + 371 pages

Already the subject of considerable pre-publication discussion, this magisterial work by one of the nation s leading neuroscientists presents a radically new view of the function of the brain and nervous system. Its...

Models of musical communication and cognition: Marc Leman, ed., special issue, Interface 18 (1–2) (1989); 160 pages☆

The title of this volume seems to imply concern with a better understanding of the relationship between mind and music. On the whole, however, this goal has not been met; and the interested reader would probably do better to turn to a discussion of music written by Marvin Minsky in 1981. This discussion has the added advantage of providing one of the most concrete arguments for how the study of music might benefit the pursuit of artificial intelligence