Neural networks almost mimics the working of human brain.The neurons are connected by axons in human brain.Same way we have neural units in neural networks.
Neural networks consist of multiple layers.And each layer has neural units .One is input layer and one is output layer.In between them we have more layers also called as hidden layers.
The number of units in input layer is equal to the number of predictor variables.
The number of units in output layer is mostly 1.And that’s because we mostly have only one variable to predict.
The units in hidden layers are called hidden units.
All the units in the input,output and hidden layer are connected with many lines representing connection weights between two units.
These lines merge with each other to form a complex network of neural units and that’s called “Neural Network”.
Like other machine learning methods, neural networks solves a wide variety of tasks, like computer vision and speech recognition,text recognition and so on that are hard to solve using ordinary rule-based programming.
If you want to learn about fitting neural networks in R,please have a look on video tutorial.