AUTOMATIC CATEGORIZATION OF 5 SPECIES OF CYMATOCYLIS (PROTOZOA, TINTINNIDA) BY ARTIFICIAL NEURAL-NETWORK

Photomicrographs of 5 species of Cymatocylis were digitised, binarised and edited by hand to remove large debris contaminating the images. A n artificial neural network (back-propagation of error) was trained to categorise 201 of these specimens after pre-processing the data by Fo urier transformation. Of the 299 trials which were carried out, 28 % d emonstrated better than 70 % correct categorisation of the data used i n the training sets. The best performing network learned to differenti ate the training data set with an error rate of 11 %. The same network gave an error rate of 18 % when presented with previously unseen data . The results of training back-propagation of error networks are prese nted and the performance and limitations are discussed and compared wi th more classical morphometric and clustering techniques for the taxon omic separation of marine plankton. This automatic technique demonstra tes the potential of neural network pattern classifiers for addressing the difficult taxonomic task of congeneric classification and also ha s wider implications for the automatic identification of field samples of marine organisms.