Morris, Eric Allen
MetadataShow full item record
This paper presents an evolutionary method of con¯guring fuzzy inference systems, entitled FISSION (Fuzzy Inference System Setup by evolutION). The primary goal of FISSION is to provide a robust software tool applicable to a wide range of supervised training scenarios. Another aim is to produce human-interpretable fuzzy inference systems with a limited number of fuzzy sets and inference rules, which nonetheless attain a high degree of accuracy. In order to accomplish these goals, this paper draws on techniques related to genetic programming, real parameter optimization, and user interface design. User-friendly software has been developed which implements many of the ideas discussed herein, and this paper presents an application of the software to predict the rate of biosolids composting.