An Experimental Comparison of Knowledge Representation Schemes

Many techniques for representing knowledge have been proposed, but there have been few reports that compare their application This article presents an experimental comparison of four knowledge representation schemes: a simple production system, a structured production system, a frame system, and a logic system. We built four pilot expert systems to solve the same problem: risk management of a large construction project Observations are made about how the structure of the domain knowledge affects the implementation of expert systems and their run time efficiency WIT: THINK THAT IT IS NECESSARY to clarify the advantages and disadvantages of knowledge representation techniques from the expert system designer’s point of view. Barr and Feigenbaum (1981) point out that “many researchers feel that the representation of knowledge is the key issue at this point in the development of AI.” Stefik et al (1982) point out some domain characteristics that affect expert system design: large search spaces, a need for tentative reasoning, time varying and noisy data. These four characteristics are elaborated into eleven case studies, and some guidelines for expert systems construcThe authors would like to express their appreciation to Dr Edward A. Feigenbaum and H Penny Nii of Stanford University for discussing the early results of this research during their visit to the authors’ laboratory The authors also would like to acknowledge the support of Dr Jun Kawasaki, the general manager of Systems Development Laboratory, Hitachi, Ltd tion are provided. This information helps an expert system designer clarify the domain’s characteristics and develop a conceptual system design. However, little information is provided for selecting adequate techniques after the system’s function (input/output) is determined. Ennis (1982) reports her experiences with building an expert system using several tools such as EXPERT, UNITS, EMYCIN, and OPS-5. The system was designed to interpret the X-ray powder diffraction spectra of rocks to determine their constituent minerals. This article focuses on expert system building tools; however, there may be many cases where no such tools are available. This article’s purpose is to present an experimental comparison of four pilot expert systems whose knowledge representations are a simple production system, a structured production system, a frame system, and a logic system. The domain of the four systems is the risk management of large construction projects. Each system’s function (inputs/outputs) is exactly the same: to assist a project manager effectively in controlling risks as they arise during project execution. We will compare the difficulties of implementing the knowledge base and the inference engine, as well as run time efficiency. The systems were implemented on a VAX II/780 using Franz-LISP. THE AI MAGAZINE Summer 1984 29 AI Magazine Volume 5 Number 2 (1984) (© AAAI)