Togaware DATA MINING
Desktop Survival Guide
by Graham Williams
Google


Neural Network

Image ann-example

Neural network algorithms can be used for regression or classification tasks.

Neural networks (often called artificial neural networks to distinguish them from the natural kind found in humans) are a data and processing structure inspired by natural neural networks. The basic idea is to connect a collection of simple neurons into a network. Some of these nodes are identified as input nodes while others are output nodes. The input data is always numeric, perhaps requiring some transformation. The numbers are propagated through the nodes of the network, being modified as they go (multiplied by link weights, and combined with other numbers at nodes), until they pop out at the output nodes. As a classification model the variable values are provided to the input nodes and the ``answer'' pops out at the output node.

Todo: Topics include: Neural Networks; Multilayer, Feed-Forward, Neural Network; Neuron Activation functions; Learning through backpropogation.

See http://www.idiap.ch/~bengio/lectures/



Subsections
Copyright © 2004-2010 Togaware Pty Ltd
Support further development through the purchase of the PDF version of the book.
The PDF version is a formatted comprehensive draft book (with over 800 pages).
Brought to you by Togaware. This page generated: Sunday, 22 August 2010