This book describes the Common KADS methodology, the product of a family of knowledge and analysis design support (KADS) projects spanning the past two decades. The last of the book's authors has been involved in this effort since the beginning of 1983. Thus, the book is particularly interesting to those who have been following their work.
The original motivation for this work was to develop a tool kit of mod-els and methods for expert system developers. The approach was based on two key observations: The first observation was that expert systems had to simulate, to some extent, human behaviour to be understandable to their users and, in cases where they're placed experts, to integrate smoothly with the existing process in which they were introduced. The major bottleneck in the effort to make expert systems like the experts they were meant to simulate was knowledge engineering, that is, extracting domain-specific knowledge from the “expert's head” and representing it in a formal representation scheme that the system could interpret [1]. The challenge then became to facilitate the knowledge-engineering process by designing common language between the knowledge engineer and the domain expert. This challenge gave rise to the issue of the vocabulary of this common language. It could not be the domain knowledge itself because domains vary a lot. The second key observation that several research groups made independently yet almost simultaneously was that tasks were common across domains. Indeed, instances of the diagnosis task can be found in medicine, car troubleshooting, soft-ware debugging, and so on. Identifying generic, domain-independent tasks, formalizing task representation, elucidating the role of the task in eliciting domain speci?c knowledge, and standardizing the design and development of expert systems then became the major research problems of the ?eld.
The common KADS methodology, as it has evolved through the years, consists of three main components: First ,it proposes a modelling framework for representing and analyzing tasks, the domain knowledge used to perform them, the agents who perform them, and the communication aspects of tasks performed by different agents. Second, it offers a set of task templates.
The disciplines of knowledge engineering and knowledge management are closely tied. Knowledge engineering deals with the development of information systems in which knowledge and reasoning play pivotal roles. Knowledge management, a newly developed field at the intersection of computer science and management, deals with knowledge as a key resource in modern organizations. Managing knowledge within an organization is inconceivable without the use of advanced information systems; the design and implementation of such systems pose great organization as well as technical challenges. The book covers in an integrated fashion the complete route from corporate knowledge management, through knowledge ...