What is neural networking?

Answer 1

http://www.itworld.com/App/543/

Answer 2

Since you're asking in the computers section, I'm guessing you're actually asking about artificial neural networks, which are usually software constructs, rather than biological neural networks, such as exist in most nervous systems.

The idea is pretty simple. You have a set of neurons, organized into layers. The first layer takes inputs, probably gathered from the outside world. This layer passes it on to the first hidden layer, which does some test on the inputs and passes its output to the next hidden layer, and so on, until the output reaches the output layer.

The tests vary depending on exactly what sort of neurons you're looking at, but generally sigmoid units are used to output a value between 0 and 1, based on the sum of the inputs, each of which is weighted by some factor.

The part that makes all this worthwhile, aside from emulation of biological processes, is that the weights on the hidden nodes can be learned through a process called back propogation. The basic idea is that if you have an input set, and you know what you want the network to finally output, you can run the input through, find out what it did actually output, and then based on the difference run a function backwards to adjust the weights of the network so it will produce something closer to your desired result.

Because of this learning capability, ANNs are usually used to set up classification systems... you teach the network to look at your input set, and make a classification of what it is, or maybe what to do next.

Answer 3

neural networking refers to the neural pathways in a persons central nervous system, and more specifically to a persons brain. the human brain is like a million computers, all networked together on a LAN called the neural network

Answer 4

It's a group of interconnected biological neurons.