|
In this article:
- What is artificial intelligence
- Artificial Intelligence Ideologies
- History of Artificial Intelligence
What is Artificial Intelligence
Artificial intelligence or AI as it is better known can be defined as the intelligence shown by an artificial unit. Mostly, this artificial unit is a computer.
Artificial intelligence is an important matter of study under computer science and deals with learning, adaptation and intelligent behavior in machines. Artificial intelligence research aims at bringing machines neck-to-neck to the human thinking procedure and hence providing them the ability to ‘think’. The corollary to this is that machines must be able to do taks that require intelligence. Examples of such activities include controlling, planning and scheduling, the ability to answer questions of diagnostic nature and so also recognition of handwriting, speech, and faces. AI systems find their use in the field of medicine, economics, military, and engineering. Apart from this, they are also used in computer games where various decisions regarding the gameplay are made by it. This can be seen in the game of chess.
Artificial Intelligence Ideologies
There are two schools of thought under Artificial Intelligence. They are as follows:
- Conventional AI
Conventional AI refers to the use of those methods that pertain to machine learning. There basic characteristics are formalism and statistical analysis. Conventional AI may also be called symbolic AI or logical AI. There are certain methods under this and have been shown below:
- Expert Systems
Expert systems are those that try to reason things in order to reach a conclusion. Such an expert system has the ability to process very large amounts of given information and they provide conclusions on that basis.
- Behavior Based AI
Behavior based AI is a modular method used for making AI systems by hand.
- Bayesian Networks
- Case Based Reasoning
- Computational Intelligence
Computational Intelligence deals with the iterative development of machines which is known as learning. Learning, as is known, is based on empirical data. Other names for computational intelligence are non-symbolic AI and soft computing. Some of the methods used under computational intelligence are:
- Neural Networks
Neural networks are systems that have a very strong ability recognize patterns.
- Fuzzy Systems
Fuzzy systems use several techniques that help them in reasoning under uncertainty. Theyy have found a widespread use industrial and consumer product control systems in the modern day world.
- Evolutionary Computation
Evolutionary computation makes use of concepts influenced from biology. These include populations, mutation and the concept of survival of the fittest to generate better solutions to the given problem. Evolutionary computation branches off into evolutionary algorithms and swarm intelligence.
Hybrid intelligent systems try to bring these two groups together. Neural networks are used to produce expert inference rules. Similarly, production rules are generated from statistical learning.
A new approach called intelligence amplification strives to achieve artificial intelligence through an evolutionary development process.
|
