Knowledge Representation for Expert Systems: A Survey and Evaluation of Techniques

Knowledge representation plays a key role in the development of any Artificial Intelligence based system. A good representation can significantly shorten development time and execution speed, while a poor representation can doom a project.

Four representation techniques are commonly used to model knowledge in expert systems: logic, production rules, semantic networks, and frames. This paper describes the application of each of these techniques in modelling mechanical systems. Advantages and disadvantages for each of these techniques are presented.