This text serves as a cookbook for neural network solutions to practical problems using C++. It will enable those with moderate programming experience to select a neural network model appropriate to solving a particular problem, and to produce a working program implementing that network. The book provides guidance along the entire problem-solving path, including designing the training set, preprocessing variables, training and validating the network, and evaluating its performance. Though the book is not intended as a general course in neural networks, no background in neural works is assumed and all models are presented from the ground up. The principle focus of the book is the three layer feedforward network, for more than a decade as the workhorse of professional arsenals. Other network models with strong performance records are also included. Bound in the book is an IBM diskette that includes the source code for all programs in the book. Much of this code can be easily adapted to C compilers. In addition, the operation of all programs is thoroughly discussed both in the text and in the comments within the code to facilitate translation to other languages. Content: Front Matter, Page iii Copyright, Page iv LIMITED WARRANTY AND DISCLAIMER OF LIABILITY, Page v Dedication, Page vii Preface, Pages xvii-xviii 1 - Foundations, Pages 1-14 2 - Classification, Pages 15-22 3 - Autoassociation, Pages 23-45 4 - Time-Series Prediction, Pages 47-66 5 - Function Approximation, Pages 67-76 6 - Multilayer Feedforward Networks, Pages 77-116 7 - Eluding Local Minima I: Simulated Annealing, Pages 117-134 8 - Eluding Local Minima II: Genetic Optimization, Pages 135-164 9 - Regression and Neural Networks, Pages 165-171 10 - Designing Feedforward Network Architectures, Pages 173-185 11 - Interpreting Weights: How Does This Thing Work?, Pages 187-199 12 - Probabilistic Neural Networks, Pages 201-222 13 - Functional Link Networks, Pages 223-230 14 - Hybrid Networks, Pages 231-244 15 - Designing the Training Set, Pages 245-252 16 - Preparing Input Data, Pages 253-278 17 - Fuzzy Data and Processing, Pages 279-326 18 - Unsupervised Training, Pages 327-341 19 - Evaluating Performance of Neural Networks, Pages 343-360 20 - Confidence Measures, Pages 361-387 21 - Optimizing the Decision Threshold, Pages 389-401 22 - Using the NEURAL Program, Pages 403-421 Appendix, Pages 423-477 Bibliography, Pages 479-490 Index, Pages 491-493
