Expert computer systems

Ordinary computer programs organize knowledge on two levels: data and program. Most expert computer systems, however, organize knowledge on three levels: data, knowledge base, and control. Computers organized in this way are often called knowledge-based svstems. On the data level is declarative knowledge about the particular problem being solved and the current state of affairs in the attempt to solve the problem. On the knowledge-base level is knowledge specific to the particular kind of problem that the system is set up to solve. This knowledge is used by the system in reasoning about the problem and is often given in the form of operators, or "pattern-invoked programs." One, many, or no operators may be applicable to the problem at any one time. If applied, an operator produces changes in the data. In the control structure is a computer program that makes decisions about how to use the specific problem-solving knowledge. Decisions are made, for example, about which operators to apply and how to apply them. This article discusses the techniques used in expert systems on each of these levels. Since these systems use a combination of artificial intelligence (AI) problem-solving and knowledgerepresentation techniques, information on these areas is also included.