Validator, A Tool for Verifying and Validating Personal Computer Based Expert Systems

Jafar, Musa; Bahill, A. Terry

doi:10.1007/978-94-009-2203-7_15

Cited by 4 publications

(2 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Based on such approaches, many automated tools have been developed and used to check for the important types of faults in different domains. These tools include algorithms that check for known potential faults, such as ESC (Expert System Checker) [Cragun and Steudel 1987]; CHECK [Nguyen et al 1985]; ARC (ART (Automated Reasoning Tool) Rule Checker) [Nguyen 1987]; RCP (Rule Checking Program) [Suwa et al 1982[Suwa et al , 1984; Validator [Jafar and Bahill 1990]; and COVER [Preece 1989]. Most tools only address a local verification of a rule base.…”

Section: Related Workmentioning

confidence: 99%

“…During the execution of a C 3 I system, faults in a rule set such as circularity and inconsistency can decrease the system performance and even cause more serious failures [Nazareth 1989]. Preece [1991] presents a set of fault definitions that can be detected by using verification programs that appeared earlier in several papers [Suwa et al 1982[Suwa et al , 1984Nguyen 1985Nguyen , 1987Cragun and Steudel 1987;Miller et al 1986;Preece 1989;Jafar and Bahill 1990]. These traditional verification programs used to detect faults in a rule base compare two rules at once (i.e., at a local level [Agarwal and Tanniru 1992]).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Algorithms to detect chained-inference faults in information distribution systems

Hwang¹,

Rine

2001

Proceedings of the 2001 ACM Symposium on Applied Computing

View full text Add to dashboard Cite

There has previously been considerable work in the quality assessment of critical control systems, such as Command, Control, Communication, and Intelligence (C 3 I) systems used in the battlefield, to verify and validate system knowledge bases and to check them for completeness and consistency. During the execution of a C 3 I system, structural faults in a rule set, such as inconsistency, can decrease the implemented system's performance and even cause more serious failures, such as performing unexpected actions. Traditional ways to detect faults in a rule set include comparisons of two rules at once. However, faults could be introduced by rule inferences such that, in the implemented system, one or more rules will be fired due to another fired rule. This paper presents newly researched and developed algorithms that can effectively and efficiently detect chained-inference faults in rule sets of information distribution system (IDS), which is a subsystem within a C 3 I system. A new directed graph paradigm called Transition-Directed Graph (TDG), used to represent IDS rule sets at nodes of the IDS, is presented. In this research, there are six categories of chained-inference rule faults defined using a TDG presentation. Based on these definitions of faults, algorithms used to detect such faults have been researched and developed.

show abstract

Section: Related Workmentioning

confidence: 99%