A COMPARISON OF RADIAL BASIS FUNCTION AND BACKPROPAGATION NEURAL NETWORKS FOR IDENTIFICATION OF MARINE-PHYTOPLANKTON FROM MULTIVARIATE FLOW-CYTOMETRY DATA

Two artifical neural network classifiers, the well-known Multi-layer P erceptron (MLP) (also known as the 'backpropagation network'), and the more recently developed Radial Basis Function (RBF) network, were eva luated and compared for their ability to identify multivariate flow cy tometric data from Jive North Sea plankton groups (Dinoflagellidae, Ba cillariophyceae, Prymnesiomonadida, Cryptomonadida, and other flagella tes). RBF networks generally performed similarly to MLPs, and slightly better in cases where the data M!ere markedly multimodal; RBF network s also have much shorter training times. The performance of MLPs was i mproved greatly by the use of a symmetrical bipolar 'transfer function ' as opposed to the commonly-used asymmetric form. The issues of netwo rk optimisation and computational efficiency in use are discussed.