Abstract:
Abstract. The knowledge level hypothesis formulated by Newell states the existence of knowledge independent of its representation. After a careful study of its implications, it turns out that the knowledge modelling theory, the knowledge representation techniques, and the knowledge acquisition process, involved in the overall knowledge engineering process, are nevertheless strongly interrelated. Thus, the claim is that the construction of knowledge based systems (KBS) at the knowledge level requires a coherent framework, which takes into account these interactions. In this article, such a framework including a knowledge modelling paradigm, the task oriented modelling, an architecture based on a high level representation language, the task model formalism, and a knowledge acquisition method, the task oriented acquisition method, is proposed. The two most significant knowledge engineering approaches, KADS and Components of Expertise, are first surveyed and compared according to these three dimensions: modelling, representation, acquisition. Then, the Task Model perspective of knowledge engineering and its implications for Architecture and Knowledge Acquisition are presented.
Citations
|
602
|
The knowledge level
– Newell
- 1982
|
|
251
|
Heuristic classification
– Clancey
- 1985
|
|
42
|
Towards a functional architecture for intelligence based on generic information processing tasks
– Chandrasekaran
- 1987
|
|
37
|
Generic tasks as building blocks for knowledge-based systems: The diagnosis and routine design examples
– Chandrasekaran
- 1988
|
|
15
|
Model-K: Prototyping at the knowledge level
– Karbach, Vo��, et al.
- 1991
|
|
10
|
Reusability and configuration of applications by non-programmers
– Steels
- 1992
|
|
9
|
The CRYSALIS Project: Hierarchical Control of Production Systems
– Terry
- 1983
|
|
4
|
Knowledge Acquisition Based on Explicit
– Linster
- 1992
|
|
4
|
Task Centered Representation for expert systems at the Knowledge Level
– PIERRET-GOLBREICH
- 1991
|
|
3
|
BB1: An Architecture for Blackboard Systems that Control, Explain, and Learn About Their Own Behavior
– HAYES-ROTH
- 1984
|
|
3
|
An Investigation of the Roles of Problem-Solving Methods
– Punch, Chandrasekaran
- 1993
|
|
2
|
generic Tasks in knowledge-basedreasoning : the right level of abstraction for knowledge acquisition
– Bylander
- 1988
|
|
2
|
A meta-interpretation model for handling mixed knowledge representations
– PRIMIO, WITTUR, et al.
- 1987
|
|
2
|
KARL : A Knowledge Acquisition and Representation
– FENSEL, ANGELE, et al.
- 1991
|
|
2
|
Hypothse du "Knowledge Level" : thorie et pratique
– GANASCIA
- 1991
|
|
2
|
Flexibility and efficiency in Blackboard Systems: studies and achievemens
– LASRI, MATRE
- 1989
|
|
2
|
KADS : A modelling approach to knowledge engineering, Esprit project P5248 kads-II
– WIELINGA, SCHREIBER, et al.
- 1991
|
|
1
|
Wielinga , B.Models of expertise in Knowledge Acquisition
– Breuker
- 1989
|
|
1
|
Die KI-Werkbank BABYLON
– CHRISTALLER, PRIMIO, et al.
- 1989
|
|
1
|
A Generic Blackbord Development System
– CORKILL, GALLAGHER, et al.
- 1986
|
|
1
|
ML) : Aformal language for KADS models of expertise.in Knowledge Acquisition
– HARMELIN, BALDER
- 1992
|
|
1
|
Organisation du contrle dans les architectures de blackboard
– LASRI, MATRE
|
|
1
|
Blackboard Systems Part I, AI magazine, 7 (2
– NII
- 1986
|
|
1
|
TASK : Task Architecture for Structuring Knowledge
– PIERRET-GOLBREICH, DELOUIS
- 1990
|
|
1
|
Vers une nouvelle architecture pour l'explicitation du raisonnement au niveau connaissances
– PIERRET-GOLBREICH
- 1991
|
|
1
|
Acquisition des connaissances orientes modles de tches
– PIERRET-GOLBREICH
- 1992
|
|
