Information Technology | Economical IT » Artificial Intelligence Notes

Source: http://www.doksinet Artificial Intelligence February 20 – February 24 • Introduction to Artificial Intelligence (AI): o Artificial intelligence (AI) is the science of making machines imitate human thinking and behavior (p. 189) o A robot is a mechanical device equipped with simulated human senses and the capability of taking action on its own (in contrast to a mechanical device that requires direction from a person) (p. 189) o • Categories of AI:  Expert systems  Neural networks  Genetic algorithms  Intelligent agents Expert Systems: o An expert system (a.ka knowledge-based system) is an AI system that applies reasoning capabilities to reach a conclusion (p. 190) o Expert systems are based on the know-how of experts in the field. This expertise is built into the system and thus does not require as much knowledge to use as a typical DSS (p. 191). o o Expert systems are designed to work at two of the phases of decision making (p. 190): 

Intelligence phase – what’s wrong; identify problem or opportunity  Choice phase – what to do; decide what to do. Most expert systems are built on the concepts of questions and rules. The expert system asks a question. If it is answered “yes”, another question appears If it is answered “no”, a different question appears. Based on the answer to this question, another question is asked. This process of question and answer continues until a decision is reached (p 192) o When developing an expert system, there are several terms to which you must be aware.  An expert system is usually built for a specific application called a domain (p. 190).  Domain expertise is the core of the expert system because it contains the steps to reach a decision (p. 194) Source: http://www.doksinet  A domain expert is the person who provides the domain expertise (p. 194)  A knowledge engineer is the IT person who coverts the domain expertise into an expert system (p.

194)  Once the knowledge engineer has converted the domain expertise into rules, the knowledge base is used to store the rules of the expert system (p. 194)  The inference engine is the part of the expert system that takes your answers and decides what to ask next (p. 194)  The explanation module is the part of the expert system that provides the reason why a conclusion was reached (p. 195) o • Problems with Expert Systems:  Converting the domain expertise into a knowledge base may be too difficult.  The expertise may be too complex to be used in an expert system.  The expert system has no common sense. Neural Networks: o A neural network simulates the human ability to classify things without taking prescribed steps leading to the solution. A neural network is an AI system that is capable of finding and differentiating patterns (p. 196) o Neural networks are most useful for identification, classification, and prediction when a vast amount of

information is available. By examining many, many examples, it determines important relationships and patterns in the information (p. 196) o In an expert system, you input hundreds, or thousands, of examples into a neural network. The neural network examines this input in many different ways until it finds an “average” solution (p. 198) o The difference between an expert system and a neural network is that an expert system is rigid and unchanging and a neural network can learn and change “on the fly” (p. 198) o The big problem with neural networks is that so much of their processing takes place behind the scenes, it is hard to relate how the solutions are found (p. 199) • Genetic Algorithms: o A genetic algorithm is an artificial intelligence system that mimics the evolutionary, survival-of-the-fittest process to generate increasingly better solutions to a problem (p. 199). Source: http://www.doksinet o In other words, a genetic algorithm finds the combinations

of inputs that give you the best outputs (p. 199) o Genetic algorithms are best suited to decision-making environments in which thousands, or perhaps millions, of solutions are possible (p. 200) • Intelligent Agents: o An intelligent agent is software that assists you, or acts on your behalf, in performing repetitive computer-related tasks (p. 202) o Types of intelligent agents:  A buyer agent (a.ka shopping bot) is an intelligent agent on a Web site that helps you, the customer, find the products and services you want (p. 202) • Shopping bots make money by selling, advertising, conducting special promotions in cooperation with merchants, or charging click-through fees.  A user agent (a.ka personal agent) are intelligent agents that take action on your behalf. Examples of tasks performed by user agents include (p 203):  • Check your e-mail • Assemble customized news reports for you • Find information on a subject of your choice Mining and

surveillance agents (predictive agents) are intelligent agents that observe and report on equipment. Examples include (p 204):  • Observing and reporting on equipment • Tracking computer networks • Watch competition and bring back price changes made by competitors A data-mining agent operates in a data warehouse discovering information. A data-mining agents detects trends in data (p. 205)