Papers on Verification and Validation of Expert Systems and Artificially Intelligent Systems

 

Validation of Expert Systems—With Applications to Auditing and Accounting Expert Systems

 

Daniel E. O’Leary

Marshall Graduate School of Business

University of Southern California (USC)

Los Angeles, CA 90089-0441

 

Decision Sciences, Volume 18, Number 3, Summer 1987

 

Abstract

 

This paper proposes a set of definitions for the concepts “validation” and “assessment” applied to expert systems (ESs).  It develops a framework for this validation and demonstrates the framework on existing accounting and auditing ESs to elicit some of the research issues involved in ES validation. 

 

Validation is critical to the design and implementation of decision making ESs.  In a setting where objectivity is sought and variance is avoided, validation ascertains what a system knows, knows incorrectly, or does not know.  Validation ascertains the systems level of expertise and investigates the theoretical basis on which the system is based.  It evaluates the reliability of the decisions made by the system.

 

The validation framework developed in this paper is research methods.  IT is designed to reflect the unique aspects of ESs (in contrast to other types of computer programs) and can be used by ES developers as a basis from which to perform validation and by researchers as a framework to elicit research issues in validation

 

Key Words: Validation, Validating, Expert Systems, Artificial Intelligence

 

 

http://www-rcf.usc.edu/~oleary/Papers/Validation%20of%20Expert%20Systems.pd.pdf

 

 

Methods of Validating Expert Systems

 

Daniel E. O’Leary

Marshall Graduate School of Business

University of Southern California (USC)

Los Angeles, CA 90089-0441

 

Interfaces, Volume 18, Number 6, November – December 1988, (pp. 72-79)

 

Abstract

 

Validating expert systems is critical.  Without appropriate validation the system may make costly errors.  There are a number of methods of validating expert systems.  They can be validated by analyzing and testing the knowledge base and inference engine individually.  However, any analysis of the system components is limited to looking at pieces of the system and not at how they work together.  They system as a whole can be analyzed by viewing it as a black box and determining the quality of its decisions or by opening up the black box and determining why it made particular decisions.

 

Key Words: Validation, Verification, Expert Systems, Artificial Intelligence

 

http://www-rcf.usc.edu/~oleary/Papers/Validating-Expert-Systems-Interfaces.pdf