Neural network and it’s related terms

What is meant by Neural network? How many types of neural networks are there? The information about it’s related terms are to be known at the present world. Who is the father of modern neuroscience? Neural network process is similar to which process? These are some of the questions one may encounter when learning about neural networks.

A Neural network is a series of algorithms that recognize relationships in a set of data through a process that mimics the way human brain operates. It is a collection of small computing units called neurons that take incoming data and learn to take decisions over time.

The first picture of neuron was drawn by Santiago Ramon Y Cajal. If we rotate the picture 90 degree drawn by him is similar to artificial neural networks. He is known as the father of modern neuroscience. Neural network process is similar to human brain process.

HUMAN BRAIN PROCESS:- The dendrites receive electrical impulse which carry data from sensors of other adjoining neurons to the Soma. In the nucleus is processed by combining them together & passed on to axon. It carries processed data to synapse & the output becomes input to other large number of neurons.

Neuron consists of three layers such as Input, Output and hidden layers. A neural network having more than one hidden layer referred as deep neural network. Input layer forward input values to the next layer. Activation function determines how a node responds to its input functions.

Perceptrons is a simplest and oldest type of neural network. They are single layered consisting of input nodes connected directly to the output node.

Convolutional Neural Network:- It is a math operation, where a function is applied to another function and the result is a mixture of two functions.

Recurrent Neural Network:- They perform same task for every element of a sequence with prior output feeding. It can make use of information in long sequences.

General Neural Network:- An input is processed through a number of layers and an output is produced with assumption that two successive inputs are independent of each other, but that may not hold in true in certain scenarios.