On the representation and the impact of reliability on expert system weights

doi:10.1016/S0020-7373(88)80071-3

International Journal of Man-Machine Studies

Volume 29, Issue 6, 1988, Pages 637-646

https://doi.org/10.1016/S0020-7373(88)80071-3 Get rights and content

Rule-based expert systems employ weighting schemas that associate weights with a rule. In the development of an expert system the reliability of the rules may be a critical variable. However, currently, weighting systems do not facilitate accounting for reliability. Accordingly, this paper demonstrates how to introduce reliability into one of the primary systems for weighting the rules. After the introduction of reliability a number of findings are discovered. First, small changes in reliability can lead to substantial changes in the adjusted weights. Second, when reliability is completely uncertain the weights become 0. Third, introducing reliability can change the signs of the revised weights. Fourth, it is unlikely that heuristics can be effectively used instead of an analytic approach.

References (7)

A. Kidd et al.
Man-machine interface issues in the construction and use of expert systems
International Journal of Man-Machine Studies
(1985)
D. Schum et al.
Comments on the relationship between the impact and reliability of evidence
Organizational Behavior and Human Performance
(1971)
R. Duda et al.
Model design in the prospector consultant system for mineral exploration

There are more references available in the full text version of this article.

Cited by (4)

Gartner's hype cycle and information system research issues
2008, International Journal of Accounting Information Systems
Citation Excerpt :
Michaelsen (1982) used EMYCIN which derived from the work of Buchanan and Shortliffe (1984), while Dungan used AL/X, developed from the work of Duda et al. (1979), and included a means of representing uncertainty with weights on the rules. After the first accounting systems there was still substantial research regarding various aspects of expert systems, e.g., regarding the weights (e.g., O'Leary, 1988a), verification and validation of the systems (e.g., O'Keefe et al., 1987; O'Leary, 1987), knowledge acquisition (Prerau, 1987) and explanation (Arnold et al., 2006). Barker and O'Connor (1989) presented a discussion about XCON, one of the first well-accepted expert systems in practice.
Quality assurance of knowledge-based systems
1997, Engineering Applications of Artificial Intelligence
A number of knowledge-based system quality assurance approaches, techniques and tools have been developed in the last decade. They include general principles of applying verification and validation in knowledge-based system-quality management, static analysis techniques, testing methods, and reliability assessment approaches. The static analysis techniques comprise checking consistency and coherence, verifying completeness, handling redundancy and subsumption, performing walkthroughs, formal reviews, and audits. Among the testing methods, program-based and functional testing are widely used. For reliability assessment, descriptive and modeling approaches have been applied. The paper gives an overview of these topics, and analyses the effectiveness of applying quality assurance methods to knowledge-based systems development.
Integrated knowledge-based system design and validation for solving problems in uncertain environments
1992, International Journal of Man-Machine Studies
Symbolic problem solving, specifically which knowledge-based systems (KBSs), in new and uncertain problem domains is a difficult task. An essential part of developing systems for these environments is determining whether the system is adequately and reliably solving the problem. KBSs that utilize heuristics have a development cycle not conducive to formal control and have high potential for error or incorrect characterizations of the problem they are meant to solve. A method of validating and testing such systems to increase and quantify their reliability is needed. Software engineering strategies for accessing and projecting the reliability of traditional software have been developed after years of experience with the cause and effect of errors. Since KBSs are new, methods for accessing and projecting their reliability are not as well understood. However, validation techniques from traditional software development can be applied to KBSs. Validation and testing techniques unique to KBSs can also be used to determine system reliability. In essence, tools and techniques can be used to meet the need for a legitimate, integrated approach to validation and testing of KBSs as they are developed.
Expert system verification and validation: a survey and tutorial
1993, Artificial Intelligence Review

^*: An earlier version of this paper was presented at the TIMS/ORSA meeting in New Orleans, May 1987.

View full text

On the representation and the impact of reliability on expert system weights*

International Journal of Man-Machine Studies

Organizational Behavior and Human Performance

Model design in the prospector consultant system for mineral exploration