Robust neural networks using the Radial Basis Function and the Counter Propagation compared with a trained Fuzzy Model
Ralf Wieland, Wilfried Mirschel and Stephan Wirth
Summer Computer Simulation Conference 2007 (SCSC 2007)
San Diego, California (USA), July 15-18, 2007
Abstract
In this paper two modern areas of soft computing (fuzzymodeling and artificial neural networks) are discussed. A feed forward neural network (NN), a counter propagation network (CP), a radial basis function network (RBF) and a trained fuzzy algorithmare compared for regional yield estimation of agricultural crops (winter rye, winter barley). A second training set for an approximation of soil microbial activities was used to dicuss different aspects for applications of soft computing.