z-logo
Premium
Knowledge acquisition for model building
Author(s) -
Cox Louis Anthony
Publication year - 1993
Publication title -
international journal of intelligent systems
Language(s) - English
Resource type - Book series
SCImago Journal Rank - 1.291
H-Index - 87
eISSN - 1098-111X
pISSN - 0884-8173
ISBN - 0-471-59368-0
DOI - 10.1002/int.4550080107
Subject(s) - computer science , knowledge acquisition , process (computing) , expert system , subject matter expert , domain knowledge , knowledge based systems , knowledge representation and reasoning , domain (mathematical analysis) , software engineering , legal expert system , perspective (graphical) , artificial intelligence , knowledge management , programming language , mathematical analysis , mathematics
When the purpose of a knowledge acquisition (KA) system is to acquire the knowledge needed to build an analytic model of a complex system, the structure of the model can be used to guide and streamline the KA process. Constraints on a system's structure can be used to generate an “intelligent questioning” sequence of requests for descriptive facts to minimize the burden on the expert or model‐builder supplying the program with information. Moreover, general knowledge about the system domain can be supplied as “meta‐knowledge” by an expert and used by the KA program to guide the search for specific knowledge (“facts”) about a particular system from a less expert user. This article describes a KA methodology and program developed to streamline the acquisition of descriptive information about complex reliability systems (e.g., telecommunicatio networks, computer systems, industrial processes, etc.). the methodology treats knowledge acquisition and knowledge representation as two inseparable parts of an integrated process of model building. the goal of the KA dialogue is formulated as minimizing the effort needed for the user and the machine to achieve a shared model of the system to be analyzed. Models are built by specializing and instantiating templates constructed from background “meta‐knowledge.” This perspective has several implications for dialogue‐based KA shells that support modeling of complex systems in limited domains. © 1993 John Wiley & Sons, Inc.

This content is not available in your region!

Continue researching here.

Having issues? You can contact us here