Title :
Over multiple rule-blocks to modular nets
Author :
Spaanenburg, L. ; Jansen, W.J. ; Nijhuis, J.A.G.
Author_Institution :
Dept. of Comput. Sci., Groningen Univ., Netherlands
Abstract :
Real production data are vague and irreproducible. This suggests fuzzy knowledge acquisition for an online learned combined fuzzy/neural network. The paper advocates a modular neural only network based on the injection of knowledge from a multiple rule block fuzzy specification. A typical controller takes 10-25 blocks with 10 rules on average, which after injection in a neural net can be personalized and adapted by small (typical 100) measurement sets.
Keywords :
fuzzy neural nets; fuzzy set theory; knowledge acquisition; learning (artificial intelligence); neurocontrollers; production control; uncertainty handling; fuzzy knowledge acquisition; measurement sets; modular nets; modular neural only network; multiple rule block fuzzy specification; multiple rule blocks; online learned combined fuzzy/neural network; real production data; Biological system modeling; Chemicals; Fuzzy logic; Fuzzy neural networks; Humans; Knowledge acquisition; Mathematical model; Neural networks; Process control; Production;
Conference_Titel :
EUROMICRO 97. New Frontiers of Information Technology., Proceedings of the 23rd EUROMICRO Conference
Conference_Location :
Budapest, Hungary
Print_ISBN :
0-8186-8129-2
DOI :
10.1109/EURMIC.1997.617405
