The use of statistical classifiers for the discrimination of species of the genus Gyrodactylus (Monogenea) parasitizing salmonids

摘要: This study applies flexible statistical methods to morphometric measurements obtained via light and scanning electron microscopy (SEM) discriminate closely related species of Gyrodactylus parasitic on salmonids. For the first analysis, taken from opisthaptoral hooks bars 5 gyrodactylid were derived images by SEM used assess prediction performance 4 (nearest neighbours; feed-forward neural network; projection pursuit regression linear discriminant analysis). The 2 methods, nearest neighbours a network provided perfect discrimination G. salaris other when using only single structure, marginal hook. Data full complement also tested analysis gave derjavini Mikailov, 1975 truttae Glaser, 1974. method had least misclassifications was therefore assessed further for individual hooks. Five parameters hook subset (total length, shaft sickle proximal width distal width) near salaris. therefore, larger numbers are required at level than level.