L. Studer et F. Masulli, BUILDING A NEURO-FUZZY SYSTEM TO EFFICIENTLY FORECAST CHAOTIC TIME-SERIES, Nuclear instruments & methods in physics research. Section A, Accelerators, spectrometers, detectors and associated equipment, 389(1-2), 1997, pp. 264-267
In this paper we show which elements have to be extracted from a chaot
ic time series in order to define the architecture of a forecaster. Th
e forecaster chosen here is a Neuro-Fuzzy System (NFS). This NFS is tr
ained by a supervised gradient descent algorithm. The NFS is made of a
layer of singleton inputs, a hidden layer of Gaussian membership func
tions and one output unit. Product is used for rule inference and sum
for rule composition. Output is given by a height defuzzifier. Test ca
ses based on Mackey-Glass time series are presented.