AI divides about into ii schools of thought: Conventional AI vs. Computational Intelligence (CI).

Conventional AI
This mostly involves methods nowadays classified under Machine learning, characterised by formalism & statistical analysis. This is also referred to as symbolic AI, logical AI, neat AI and Good Old Fasioned Artificial Intelligence (GOFAI). (Too look at semantics.) Methods include: Expert systems: apply intelligent capabilities to email the guide. Renowned patterns might run big numbers of known principles & provide conclusions according to the children. Case based reasoning

Computational Intelligence
This involves iterative learning of connectionist system parameter tuning, based on empirical data. This is also referred to as non-symbolic AI, scruffy AI or soft computing. Methods come: Neural networks: systems capable of finding & differentiating between system. Fuzzy logic: a system for logical thinking under uncertainty, has been widely utilized within industrial control systems. Evolutionary computation: applies biologically inspired conception like populations, mutation & survival of the fittest to generate increasingly better solutions to the problem.

Applications
Natural language processing Optical character recognition systems can translate arbitrary typed European script into text. Handwriting recognition is used around hundreds to thousands of personal digital assistants. Speech recognition is commercially available and is widely deployed. Computer vision systems are utilized around several industrial applications ranging from either hardware verification to security systems. Intrusion detection systems Game AI Program synthesis Robotics