Heirarchical classifier

Hierarchical Classifier

A hierarchical classifier is a hierarchy of individual computational units that works to classify a set on inputs to a correct class that then becomes the output. Most hierarchical classifiers will be in the form of a tree or a directed acyclic graph with the root representing the classification output and leaves at the lowest levels representing input. There are many applications of hierarchical classifiers in areas such as artificial intelligence, computer vision, automatic theorem proving, object recognition, neural computing, etc.

Theory

Much of the theory that serves as the basis of hierarchical classifiers is based on the workings of the human brain, neural networks, and statistical analysis. The human brain has been shown to build up the complex ideas it is capable of processing from very tiny individual units of data that are inputted by sensory perceptors from the outside world. All evidence points to a hierarchical structure of processing where the base data is abstracted many times to achieve concepts about the outside world. In vision,for example, these concepts represent objects, or a decription of a scene. The brain must abstract to a sufficient extent such that