Avtor/Urednik		Rovetta, G
Naslov		Perspectives in Lyme borreliosis
Tip		članek
Vir		Acta Dermatovenerol Alp Pannon Adriat
Vol. in št.		Letnik 3, št. 1-2
Leto izdaje		1994
Obseg		str. 133-6
Jezik		eng
Abstrakt		The principles of an automatic pattern recognition and classification methods are explained. A neural network (NN) is a computing structure, its purpose is to get input data and return some kind of output. This task is not performed by means of an algorithm, but rather learnt by the network itself by looking at a certain number of examples (training set). NN offers the possibility of being trained by examples and exhibits a remarkable generalization power that lets these data to be correctly interpreted. Several applications of such systems have confirmed that they can be used to establish the diagnosis and to assess the prognosis.
Deskriptorji		LYME DISEASE NEURAL NETWORKS (COMPUTER) PATTERN RECOGNITION