Knowledge acquisition with fuzzy modeling
Tapio Rauma · Proceedings of IEEE 5th International Fuzzy Systems · 2002
The spiral model for the development process is found to be the most suitable approach in constructing knowledge-based expert systems. One possible pitfall of using the spiral model is the number of rounds needed in developing an expert system. The pitfall can be avoided by using as reliable as possible methods in knowledge acquisition. Unfortunately there seem to be few methods for validating expert knowledge. In this study we present methods to validate the acquired knowledge by using a fuzzy process model. A fuzzy process model is built using the expert knowledge, and the knowledge is analyzed using data measured from a real process and the FLS tuning method.