Artificial Intelligence Artificial intelligence is an advanced branch of science which enables digital computers to perform complex tasks that are usually associated with requiring human intelligence. These tasks can vary from game playing to robotics. The best way to describe A.I is by comparing it to a human brain, digital computers consist of a network of switches and computers make decisions by using on/off switches. The basic element of the brain, the neuron also uses two state operations. This means that the human thought process can be linked to transistors within a computer. One definition of Artificial intelligence is “The science of making machines do things that require intelligence if done by men” (Marvin Minskey). The important word in this definition is INTELLIGENCE as there can be different perceptions of intelligence such as the ability to learn and acquire knowledge or just plain common sense. Some other researchers believe that A.I is just the duplication of human thought without any intelligence.
Artificial intelligence is different from normal programming by the way it solves a problem. A normal programming language such as BASIC must use a step by step method to solve a problem. This normally means that you will have to flow chart the solution then translate it into the BASIC programming language that carries out the solution. This step by step programming is called algorithmic programming. An algorithm is a well defined set of rules to solve a problem in steps. Once you have developed an algorithm the computer will take a precise amount of time to find the solution. A.I programming is different as it uses the common sense riles of thumb to solve problems just like humans do. In order to do this A.I programming must define and represent problems within the computer so that the application of certain rules will eventually lead to a solution if there is one. A.I must also provide means of directing the search for a solution to the problem using existing knowledge and common sense rules of thumb.
Knowledge based systems are system that use Artificial intelligence tools when problem solving so that it can support human decision making. Knowledge based systems do this by having a massive database of expert knowledge. These systems are used to help support humans in problem solving and can even be used for training as well. In some cases KBS are more efficient than humans as they have qualities such as efficiency and effectiveness. Society and industry are becoming reliant on Knowledge based systems as they can act as expert decision making which allows businesses to function at a higher level of consistency. KBS are capable of understanding the information being processed and make a decision based on it. KBS are different to normal systems as normal systems do not understand the information being processed. KBS are an example of fifth generation computer technology and its main objectives are to provide a high level of intelligence and offer a vast amount of knowledge in different areas. Human beings can learn information and then forget it when not in regular use, simulation of this is important on knowledge based systems. A KBS can be manually updated or automatically updated by machine learning. “There are five main types of knowledge based systems, expert systems, linked systems, CASE based systems, database in conjunction with an intelligent user interface, intelligent tutoring systems” (Tuthhill and Levy 1991).
An artificial neural network is a mathematical invention that was inspired by the study of biological neurons. The neurons are the basic signalling units of the nervous system; they have dendrites which branch out to other neurons to receive signals to each other. Artificial neural networks are similar to this as they consist of a pool of simple processing units which communicate be sending signals to each other over a large number of connections. Artificial neural networks are used to perform tasks that a computer with a normal processor cannot perform. An advantage of a neural network is that it learns and does not have to be reprogrammed. The best way to describe an artificial neural network is by mapping an input space to an output space. Most of the time artificial neural networks are nonlinear systems that learn to perform a function from data.
There are many different examples of Artificial intelligence applications a very common one is computer games. Since the creation of A.I people have been testing themselves against the computer. One game that is a good example of this is chess the A.I will play well against humans by using brute force computing, looking at hundreds and thousands of different positions. Another good example of A.I is robotics. A robot uses A.I by sensing its surroundings and responding to a changing environment just like humans do. This is needed when robots are used in factories to perform operations such as parts assembly and spray painting, if a certain part has run out or an object is there which shouldn’t be the A.I will respond to this and try to solve the problem. Expert systems are another example of A.I these systems are used to put someone’s information into a fixed set of categories such as credit card application the A.I will then find information about the person’s previous credit rating. One final example of an A.I application is speech recognition this can be used to give computers certain instructions and also used for security as it will only respond to a certain voice.
Bibliography
Robotics and A.I: An introduction to applied machine intelligence, Andrew C. Staugaard, JR. Engineering Artificial Intelligence Software, Derek Partridge. Knowledge Based Systems, Spyros G tzfesta. Www.knowledge-based-system.html. www.aaai.org/aitopics/pmwiki/pmwiki.php/AITopics/Applications