Artificial Intelligence (AI) is the area of computer science that focuses on creating machines that are capable of simulating human intelligence. John McCarthy who coined the term in 1956 defining it as "the science and engineering of making intelligent machines." Artificial Intelligence is usually associated with Computer Science, but it is also used in other areas of study such as Math, Psychology, and Biology. Our ability to combine knowledge from all these fields will ultimately benefit our progress in the quest of creating an intelligent artificial being.
The four main categories of artificial intelligence are:
-Expert system - a computer program that simulates the judgement and behavior of a human that has expert knowledge and experience in a particular field. These programs mimic the reasoning process that an expert would follow when solving difficult problems in a particular field.
Expert System (on the class site)
-Neural network - a computer model that tries to copy a biological neural model of the human brain. It is an adaptive model that is structured to function in the same way that a human brain would.
-Genetic Algorithm - A search procedure that is based upon the mechanics of natural selection and genetics, that seeks to continually produce better solutions to a problem.
-Intelligent Agent - a program that gathers information or performs a service without a users immediate presence and on a regular schedule (ex. an alarm system). An agent program, using specifications that the user has provided it with, searches all or certain parts of the Internet, gathers information the user would be interested in, and presents the information on a periodic basis.
Evaluating artificial intelligence:
In order to determine if an agent is intelligent, Alan Turing creating the Turing Test to test all major problems of artifical intelligence.
Alan Turing's article "Computing Machinery and Intelligence" described the certain conditions that must be present in order to classify a machine as intelligent. He argued that if a machine is considered intelligent if it could successfully pretend to be human to an observer administering the test. The observer would interact with the machine and a human, and have to determine who was actually human and who was machine.
According to Turing there are four classes of outcome for an AI test: