NEURO NETWORKS

NEURO NETWORKS The topic "The Neural Networks" is to introduce all the basics of an emerging technology: neural networks - their function, generic structure, terminology, types and uses. The components or parts of the neural networks-- processing units, weighted interconnections, activation-rule based on which the neural networks work and finally the last component, which is quite often an optional one, the learning rule. The characteristics of the neural networks - adaptive learning, self organiztion, fault tolerance, parallel computing have been highlighted. The comparison of neural networks with conventional computers, indicating the advantages of neural networks over conventional computers,is made. The similarities between artificial neuron and biological neuron, based on the concept took its shape, is then presented. Four architectures of neural networks exist and each of them can be studied. The learning process, which is the determination of weights by the neural networks. The behavior of a neural NETWORK depends on the weights as well as the input-output function. The report finally throws the light on the application of the neural networks in various fields. The real time example of the neural networks is then demonstrated. Like any other technology, a neural network too has its own set of advantages and disadvantages which is at-last elucidated.