15-09-2017, 03:21 PM
In information technology, a neural network is a system of hardware and / or software modeled after the operation of neurons in the human brain. Neural networks - also called artificial neural networks - are a variety of deep learning technologies. Commercial applications of these technologies generally focus on solving complex signal processing or pattern recognition problems. Examples of significant business applications since 2000 include handwriting recognition for check processing, voice-to-text transcription, oil exploration data analysis, time prediction and facial recognition.
A neural network usually involves a large number of processors that operate in parallel and arranged in levels. The first level receives raw input information - analogous to the optic nerves in human visual processing. Each successive level receives the output of the preceding layer, rather than the raw input - in the same way that the neurons furthest from the optic nerve receive signals from the nearest ones. The last level produces the output of the system.
Each processing node has its own small sphere of knowledge, including what it has seen and the rules with which it was originally programmed or developed for itself. The levels are highly interconnected, which means that each node at level n will be connected to many nodes at the n-1 level - its inputs - and at the n + 1 level, which provides information for those nodes. There may be one or more nodes in the output layer, from which you can read the response it produces.
Neural networks are characterized by being adaptable, meaning that they change as they learn from initial training and subsequent careers provide more information about the world. The most basic learning model focuses on the weighting of input streams, which is how each node ponders the importance of input from each of its predecessors.
A neural network usually involves a large number of processors that operate in parallel and arranged in levels. The first level receives raw input information - analogous to the optic nerves in human visual processing. Each successive level receives the output of the preceding layer, rather than the raw input - in the same way that the neurons furthest from the optic nerve receive signals from the nearest ones. The last level produces the output of the system.
Each processing node has its own small sphere of knowledge, including what it has seen and the rules with which it was originally programmed or developed for itself. The levels are highly interconnected, which means that each node at level n will be connected to many nodes at the n-1 level - its inputs - and at the n + 1 level, which provides information for those nodes. There may be one or more nodes in the output layer, from which you can read the response it produces.
Neural networks are characterized by being adaptable, meaning that they change as they learn from initial training and subsequent careers provide more information about the world. The most basic learning model focuses on the weighting of input streams, which is how each node ponders the importance of input from each of its predecessors.