Artificial neural network is a computation framework used to realize pattern recognition on certain subject. Artificial neural network is built from computational nodes which are interconnected with each other like a large web. The network are used to map large numbers of values into several classification labels, the network structure together with the parameters of the network can be viewed as a mapping from observed data into a label, in which the label can be used to correctly describe which does these observed data belongs to.
Artificial neural network is used to make predictions regarding a classification problem. Typical cases of the classification problem are facial recognition, ..., etc.