<XML><RECORDS><RECORD><REFERENCE_TYPE>3</REFERENCE_TYPE><REFNUM>6811</REFNUM><AUTHORS><AUTHOR>Kocijan,J.</AUTHOR><AUTHOR>Banko,B.</AUTHOR><AUTHOR>Likar,B.</AUTHOR><AUTHOR>Girard,A.</AUTHOR><AUTHOR>Murray-Smith,R.</AUTHOR><AUTHOR>Rasmussen,C.E.</AUTHOR></AUTHORS><YEAR>2003</YEAR><TITLE>A case based comparison of identification with neural networks and Gaussian Process models</TITLE><PLACE_PUBLISHED>IFAC International Conference on Intelligent Control Systems and Signal Processing Faro, Portugal, April 08-11, 2003 </PLACE_PUBLISHED><PUBLISHER>International Federation of Automatic Control</PUBLISHER><LABEL>Kocijan:2003:6811</LABEL><ABSTRACT>In this paper an alternative approach to black-box identification of non-linear dynamic systems is compared with the more established approach of using artificial neural networks. The Gaussian process prior approach is a representative of non-parametric modelling approaches. It was compared on a pH process modelling case study. The purpose of modelling was to use the model for control design. The comparison revealed that even though Gaussian process models can be effectively used for modelling dynamic systems caution has to be exercised when signals are selected. </ABSTRACT></RECORD></RECORDS></XML>