Neural Network Data Analysis Using Simulnet™, Softcover reprint of the original 1st ed. 1998

This book and software package complements the traditional data analysis tools already widely available. It presents an introduction to the analysis of data using neural network functions such as multilayer feed-forward networks using error back propagation, genetic algorithm-neural network hybrids, generalised regression neural networks, learning quantizer networks, and self-organising feature maps. In an easy-to-use, Windows-based environment it offers a wide range of data analytic tools which are not usually found together: genetic algorithms, probabilistic networks, as well as a number of related techniques that support these. Readers are assumed to have a basic understanding of computers and elementary mathematics, allowing them to quickly conduct sophisticated hands-on analyses of data sets.

Scope of this Text.- What Is Expected from the Reader.- An Outline.- Computer Requirements.- 1 The Simulnet Desktop.- Desktop Components.- 2 Data Analysis.- The Substantive Question.- Neural Network Analysis.- Genetic Algorithms and Neural Networks.- The Probabilistic Network.- The Vector Quantizer Network.- Assessing the Significance of Network Results.- Network Application Examples.- Fractal Dimension Analysis.- Fourier Analysis.- Eigenvalue Analysis.- Coherence and Phase Analysis.- Mutual Information Analysis.- Correlation and Covariance Analysis.- 3 Acquiring and Conditioning Network Data.- Data Specification.- Data Collection.- Data Inspection.- Data Conditioning.- Detrend—Order 0.- Standardize Columns.- Frequency Filtering.- Principal Component Analysis.- Principal Component Data Reduction.- 4 A Data Analysis Protocol.- A Preprocessing Checklist.- Analyzing Experimental Data.