This paper is an excellent introduction to frame-based representations. No previous background in artificial intelligence is needed to comprehend the paper. It is tutorial in nature, describing the advantages of a frame-based representation over other alternatives for representing information in a knowledge-based system (expert system). The authors use KEE from IntelliCorp to illustrate how a framed-based system operates. The only thing missing from the paper is an attempt to point out the weaknesses of frame-based representations.

The paper explains how frames and their associated slots may be used to represent concepts, classes of objects, or individual instances of objects. The authors point out that one of the advantages of frames is that some inferencing mechanisms are built in, such as exploiting the inheritance hierarchy of frames. Another advantage is that arbitrary procedures may be attached to slots. One of the uses for these procedures is that they may act as demons to simulate sensors or monitors. Examples are given to show how KEE may be used with production rules to give the knowledge engineer the best of both worlds to build his knowledge-based system. The paper concludes with an example of a real world application which utilizes the integration of the frame-based and production-rule-based representations.