Selecting expert system frameworks within the Bayesian theory

  • Authors:

  • S W Norton

  • Affiliations:

  • Siemens Research and Technology Laboratories, 105 College Road East, Princeton, NJ and PAR Government Systems Corporation, New Hartford, NY

  • Venue:
  • ISMIS '86 Proceedings of the ACM SIGART international symposium on Methodologies for intelligent systems
  • Year:
  • 1986

Quantified Score

Hi-index 0.00

Visualization

Abstract

Because of the impact that the choice of an expert system framework has on the success of expert system projects, it is important for anyone evaluating expert systems and expert system tools, whether casually or critically, to consider a variety of issues. As background, this paper provides a general description of the expert systems paradigm, and detailed descriptions of two expert system frameworks for reasoning under uncertainty. Although both of the expert systems described have roots in the Bayesian theory, their differences are significant enough to fuel a discussion of the most significant issues in evaluating the applicability of particular expert system frameworks. We cover knowledge representation and knowledge structures, the initial phases of knowledge engineering, tuning and validating the knowledge base, as well as utilization of the finished expert system.