Samarasinghe S. Neural Networks for Applied Sciences and Engineering. From Fundamentals to Complex Pattern Recognition

Auerbach Publications, 2006, -582 p.This book is an exploration of neural networks for pattern recognition in scientific data. An important highlight is the extensive visual presentation of neural networks concepts throughout. This book is motivated by the necessity for a text that caters to both researchers and students from a wide range of backgrounds, one that puts neural networks into a multidisciplinary scientific context. For the last seven years, I have taught neural networks to graduate students from diverse backgrounds, including biology, ecology, applied sciences, engineering, computing, and commerce at Lincoln University in New Zealand. My interactions with these students evolved my presentation of the material in such a way that it makes networks and their internal details transparent, thereby building confidence in the methods. Visual presentation became an invaluable tool in making difficult mathematical concepts easier to grasp. This book is a reflection of these efforts and of my own interest in exploring neural networks.
My intent is to provide a sound theoretical background within an applied context. My experience has shown that learning combined with hands-on applications using neural networks software provides the best outcome. Additionally, practical tutorial sessions to complement the theoretical treatments have been very successful in presenting this material.
I have designed this book to introduce neural networks to senior undergraduate and graduate students from applied fields of research with some mathematical and basic calculus background. Simple presentations in conjunction with visual aids make it possible to unravel a network to understand the mathematical concepts and derivations, and to appreciate the internal workings of neural networks that are considered to be a ‘black box’ by many.From Data to Models: Complexity and Challenges in Understanding Biological, Ecological, and Natural Systems
Fundamentals of Neural Networks and Models for Linear Data Analysis
Neural Networks for Nonlinear Pattern Recognition
Learning of Nonlinear Patterns by Neural Networks
Implementation of Neural Network Models for Extracting Reliable Patterns from Data
Data Exploration, Dimensionality Reduction, and Feature Extraction
Assessment of Uncertainty of Neural Network Models Using Bayesian Statistics
Discovering Unknown Clusters in Data with Self-Organizing Maps
Neural Networks for Time-Series Forecasting