Rule-based approach vs. neurocomputer approach: building a design type expert system
S. Tsutsui · 2002
Two types of class scheduling systems, one based on the traditional rule-based approach and the other on a neurocomputing approach, proposed for use at Hannan University are discussed. The class scheduling problem is first outlined briefly and the structures of both systems are described. Then, a quantitative comparison between them is carried out. The advantage of the neurocomputing approach is that it does not require knowledge acquisition, a major problem in the development of expert systems. However, the neurocomputing approach proved to be limited in representing certain valuation conditions.>