The key to artificial intelligence is data science.  Powerful computers have been created to process the huge amount of data that has been created in the world due to the advancement of information and communication technology.  Artificial intelligence uses a method to process this data that mimics the way the human brain works.  This is commonly referred to as the neural network or neural net.  A neural network is a computing system that processes a large amount of data or information, understands everything as a human being, learns from it, and gradually corrects various errors.  For example, a computer will look at a picture of a car only as a picture of a car, not as a data, such as 1 or 0.



 A typical neural network can contain millions of artificial neurons, called units.  These units are arranged in sequence with each other and each is connected to each other.  There are some units that receive different types of information, such as the human brain trying to recognize something by looking at its shape, structure, color, etc.  These units are called input units.  These help to recognize or process neural networks.  On the other side of the network is the output unit which allows the computer to express the knowledge gained from the input.  Between this input unit and the output unit lies the Hidden Unit, which is made up of most of the artificial neurons in the network.  When the neural net is trained with the input and output levels, the hidden level changes accordingly.  As a result, when a completely new input is given to the neural net at the end of the training, it is able to give the possible correct output using the hidden level.  The more data that is used to train the hidden level, the better the output will be.  If there is more than one, the neural network will become much more intelligent and even the neural net will be able to use the data itself.  This is called deep learning.