Artificial Intelligence (AI)

The capability of machines to imitate the intelligence in humans is called Artificial Intelligence(AI). It is basically the simulation of the intelligent behavior of humans in computers. Artificial Neural Network (ANN) redirects here.

Human intelligence can be defined as the mental ability for reasoning, problem-solving, memorizing, learning, understanding etc. The human brain and nervous system are responsible for the human intelligence. A system alike computer is modeled in the human brain and nervous system. Biological neurons in the nervous system process and transmit the information to the brain.

ANN is an interconnected group of artificial neurons that uses a computational model for processing information. These artificial neurons share some of the characteristics of biological neural networks. The goal of ANN is to solve problems the same way a human brain would do.

MODEL OF BIOLOGICAL NEURON

Figure 2. Model of Biological Neuron

A large number of neural cells are there in the brain which is responsible for the processing of information. Various capabilities of the brain are as a result of the massive interaction and parallel processing of these neurons. The different section of the neuron is discussed below.

• Dendrites- They are branches extending from the cell body. Dendrites receive signals from other neurons.
• Soma or cell body- Soma supports chemical processing and production of neuro-transmitters. It contains the nucleus. The incoming signals are processed and converted into output signals in the cell body or soma. The process of transmission is by diffusion of the chemical called neuro-transmitters.
• Axon- It acts as a transmission line to carry the information from one soma to another.
• Axon Hillock- All the information are gathered together and processed here. An action potential will be initiated in the hillock and propagated through the axon.
• Synapse-It is the point of interconnection between two neurons. Communication between neurons takes place at this point.
• Terminal buttons- It releases neuro-transmitters.

STRUCTURE OF ARTIFICIAL LINEAR NEURON

Figure 3. Elements of artificial neuron

• Weighting Factors (w)- Here w1, w2, ..wn represent the strength of the inputs x1, x2,...xn. That is each input will be multiplied by a weighting factor 'w'. Here I am the input. At the summation point ∑,
  
• Threshold (Ø)- It processes the input signal and produces the final output Y in the desired form. The sum of the inputs is passed to the non-linear function, that is the activation function and performs various actions to produce an output Y as produced by a human biological neuron.

CHARACTERISTICS OF ANN

• ANN can process the signals in parallel and also with great speed. All the information will be in a distributed format.
• ANN will be made to train with several examples. And these examples will be stored in the structure. Therefore they can identify objects that are previously trained.
• They have mapping capabilities.
• They are robust systems and are noise immune.
• Some other characteristics include learning, categorization, and generalization.

APPLICATIONS OF ARTIFICIAL INTELLIGENCE

• In the medical field - They can act as a clinical decision support systems for medical diagnosis. ANN can design the treatment plan. They can create companion robots for old-aged people. ANN can go deep into the medical records of the people to provide useful information.
• Industries- Robots can be introduced to the jobs that are dangerous to humans.
• Customer services- Telephonic and online customer support can be made possible with ANN.
• Programming games- Different programming games and robot alike toys are possible with artificial intelligence.