Knowledge engineering

Construction of expert systems is facilitated if it is possible (at least approximately) to describe (model) expected signal from defect and non-defect pieces. If no models for the problem are available then the knowledge about the problem has to be acquired from an expert (the NDT inspector). However, the knowledge possessed by the expert is often incomplete and not well formalised, which makes knowledge acquisition a difficult task for the knowledge engineer. 

DENDRAL proved to be fundamentally important in demonstrating how rule-based reasoning could be developed into powerful knowledge-engineering tools [93]. 

The expert s knowledge must be incorporated in the expert system in a formal way such that it is unambiguous to the system and the system can make forward and backward inferences. This requires special training and skill, and the person possessiag them is known as a knowledge engineer. 

Neural networks have the following advantages (/) once trained, their response to input data is extremely fast (2) they are tolerant of noisy and incomplete input data (J) they do not require knowledge engineering and can be built direcdy from example data (4) they do not require either domain models or models of problem solving and (5) they can store large amounts of information implicitly. 

Data on the toleranee eapability of the manufaeturing proeesses eovered were eom-piled from international standards, knowledge engineering in speeialist businesses... 

C. A. Copenhaver, Knowledge Engineering Environment (KEE) from Intellicorp, Packag. Software Rep., 15 (1989) 5-20. 

A typical feature of expert systems that support frames is inheritance. Frames can be organized in a hierarchical structure. They can inherit properties (attributes) from frames that are higher in the hierarchy. The latter are therefore called parent frame and the former child frame. There are many varieties of the inheritance principle. Frames can have only one parent frame (simple inheritance) or may have multiple parent frames (multiple inheritance). All attributes can be inherited (full inheritance) or only a few, selected by the knowledge engineer, may be inherited (partial inheritance) by the child frames. An example of a simple inheritance organization of frames is shown in Table 43.1. The frame Organic Compound is the parent frame. The frames Ester and Acids are child frames of Organic Compound . A typical example of inheritance is instantiation. The frame Acetic acid is a child of Acids and, since no extra attributes are added, it is also an instantiation. 

J. Sandberg, B. Wielinga and G. Schreiber, Methods and techniques for knowledge management what has knowledge engineering to offer Expert Syst. Applic., 13 (1) (1997) 73-79. 

Before any information can be entered into the knowledge base, decisions must be made about how the rules and data will be represented. The question of knowledge representation is a matter for the knowledge engineer who, from a preliminary discussion with the expert, and meetings with potential... 

In broad terms, the aim of knowledge engineering is to integrate human knowledge with computer systems. 

Knowledge engineering is the technology base of the "Second Computer Age"... 

Knowledge engineering is the technology behind construction of expert systems, or knowledge systems, or expert support systems. Such systems are designed to advise, inform and solve problems. They can perform at the level of experts, and in some cases exceed expert performance. They do so not because they are "smarter but because they represent the collective expertise of the builders of the systems. They are more systematic and thorough. And they can be replicated and used throughout a laboratory, company or industry at low cost. 

A Knowledge-Engineering Facility for Building Scientific Expert Systems... 

---