Artificial Neuron & its components

(AI) Artificial intelligence is new development in computer science to perform specific functions like human being. Neural theory was influenced from psychological theories of animal learning such as many animals learn many things even they know who love them or their enemy, same like artificial neuron is derived from human brain which relies on neurons it is based on the theory of automate in which intelligence is enabled into machine, in simple words it is tried to make such machine or system that works according to the situation.

For example we prepare a machine that has two functions.

1. It will work as a AC (air condition) when temperature goes above 30 degree.
2. It will work as a Heater when temperature goes below 20 degree.

For this purpose there are few questions how to tell machine what is temperature of room now, in which time Ac is to be on, because these all things have to done by machine not human being, on the other hand machine has not any type of feelings, emotional then scientists tried to make an artificial machine that relys on artificial neurons it consists numbers of components.

Components of Artificial Neuron:

Input Signal: (environmental Signal)

Artificial machine relys on data that is coming towards it to perform specific function, this data may come from any source around it such as air pressure, changing external temperature, voice affects etc. It can also come owing to activation of other neurons on the behalf of these values (air pressure, temperature etc.) then artificial neuron start working on the basic of getting input from environment.

Set of Real Valued Weight: (measurement of input Signal)

It is important to know the value of input signal, actually particular signal strength is saved in artificial device and here signal strength is measured for further function. When strength of external signal is equal to internal saved signal, system becomes starting work.

Activation Level:(Sysem is ready to work)

When a system becomes in working position that is called active mode , so neuron activation mode depends on commutative strength of signals, it is kept according to the situation when we have to place over system in noisy area activation level is adjusted that is why it is kept variable in many devices.

Threshold Function:  (On/off Situation)

There is very slight difference between activation level and threshold function when a system is on at certain point called activation level as it is activated completely this situation is called threshold, Simply when activation level achieved neuron start working, it produces on/off states of neurons.

Network Topology:     (Physical Layout)

In order to perform any function connection among neuron is basic thing, in simple words the physical layout or arrangement of connected neurons in a network called topology, it is done in parallel processing.

Learning Algorithm:  (Mathematic System)

Various kinds of algorithms are used to form artificial neurons, because an artificial system does not have any sense of emotional that is why neuron starts work by making connection each other through follow algorithms that is already saved, an artificial machine always work to the point by following algorithm.

Applications:

1. It is being used in Air Traffic control system.
2. Robots are being developed by using artificial neurons.
3. It is used in Security sensor device.
4. Expert Systems are being more expert.
5. Automatic On/Off switches.