Skip to content

Epistemological challenges for the next generation AI and expert systems

Citations of this article
Mendeley users who have this article in their library.
Get full text


Expert Systems represent a widespread newer application of Artificial Intelligence. In essence, expert systems use computer programs to accumulate the experience of experts in a given field and then provide either solutions to complex problems or explanations as aids to decision-making. These systems are particularly good at providing information related to problems of a technical nature. However, recent research and literature from the fields of AI, psychology, curriculum theory, management education, professional decision-making and the sociology of knowledge has begun to emphasize the need for approaches to knowledge which go beyond the “technical “and incorporate perspectives of a more “practical “and “critical “nature. It would suggest that designers of expert systems be aware of their epistemological assumptions—that is assumptions about the nature and structure of knowledge and how it might be acquired. This paper elaborates on both traditional and emerging conceptions of learning and knowledge and then discusses the challenges presented to designers of expert systems, as well as higher education, in incorporating forms of knowledge that go beyond the technical.




Steier, L., & Mackay, D. A. (1990). Epistemological challenges for the next generation AI and expert systems. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 451 LNAI, pp. 179–198). Springer Verlag.

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free