For practicing and student engineers, physicist Priddy and electrical engineer Keller introduce artificial neural networks without bogging down the principles in mathematics, which are presented in appendices in sufficient detail for most of the common neural network algorithms. Among the topics are learning methods, data normalization, output coding, unsupervised training methods, recurrent neural networks, a number of applications, and dealing with limit amounts of data. Annotation ©2005 Book News, Inc., Portland, OR