Faith Brown September 9, 2014

Artificial Intelligence is something that has been around for quite some time. It is simply the science of making machine imitate human thinking and behavior. There are many types of businesses and organizations that use artificial intelligence. Many of them are government organizations but are not limited to hospitals and local businesses. Artificial intelligence can be used for the simplest of things like counting inventory or to the more complex of things such as reading electrocardiograms. With artificial intelligence, we have been able to reduce the time that it takes an actual human to do something. These systems are in place so that we spend more time on the more critical things. There are four types of artificial intelligence. The four types are expert systems, neural networks (and fuzzy logic), genetic algorithms, and agent-based technologies. All of the these systems have their benefits as well as their not so good features.
For starters, we will examine expert systems. Expert systems, which is also referred to as knowledge-based systems, is an artificial intelligence system that applies reasoning capabilities to reach a conclusion. These systems are used for diagnosing problems. They also help with coming up with a solution to the problem. These systems are built for a specific domain that they will be ran on. The knowledge base for these systems contain both factual and heuristic knowledge. This means that the factual information could be found in textbooks while the heuristic information is going to be derived from experimental cases. Systems of this nature are widely used. They can be found in accounting firm for tax planning and in hospitals for diagnosing diseases. The experts found a way that the system could search the internet for the answers that it needed. The systems would gather the facts and then the computer based would give the facts or answer the questions. While expert systems are knowledge-based systems, agent-based technologies sort of work in the same manner.
Agent-based technologies are a small piece of software that acts on your behalf performing tasks assigned to it. There are five types of agent-based technologies. They are autonomous agent, distributed agent, mobile agent, intelligent agent, multi-agent systems. The two agent-based technologies that we have discussed are the intelligent agent and multi-agent systems. The intelligent agent has reasoning and learning as a form of AI incorporated into it. Intelligent agent has subgroups like information, monitoring and surveillance, and user agents (personal agents) that it uses to help with the reasoning and learning part of the AI. They will search for the information, observe what actions are being taken and report on an area of interest, as well as handle any personal matter that you may have. Another agent-based technology is data-mining agent. This agent operates in a data warehouse discovering information. Multi-agent systems learn how people-based systems behave, predict how they will behave under a given set of circumstances, and improve human systems to make them more efficient and effective. These agents have the ability to work independently but they must also work together to keep everything in order and to complete the task at hand. The downfall to this is that if one agent has provided the incorrect information then the entire would have given an incorrect answer. This is simply because each agent depends on the one prior to them to retrieve the right information and provide it to them. Everything must work together in order to complete a common task that has been given.
The next AI system is genetic algorithms. This system is devoted to the evolution or “survival of the fittest” theory. This system is in place to find the best outcome for the combinations that it has been given. Genetic algorithms is in place to do repetitive number computing. The types of computing that is done would take too long to do by actual humans. These systems are used within companies such as Boeing who has it's algorithm in place to design aircraft parts for some of their planes fan blades. Doing something of this nature on such a large object would take too many man hours so they have this algorithm that will compute the necessary information that they will need. Due to the nature of these problems that the algorithm normally handles the evaluation of these outcomes will be quick and very efficient.
The final system for artificial intelligence that is used is neural networks and fuzzy logic. Neural networks are capable of finding and differentiating patterns, while fuzzy logic is mathematically handling imprecise or subjective information. Neural networks are used in tablets, police departments, medicine, and in businesses as a part of the security features. The detection of patterns and speech recognition is a vital part of what this system does. It looks for any pattern whether it is correct or not. This is good for identification, classification, as well a predictions of what is to come. Along with that the fuzzy logic helps with finding the anomalies. It uses an approach from vague to ambiguous. This logic is a set of steps that are taken to relate everything together.
Some problems with these systems is that some take more time than others and being that we are in a fast paced environment, we need the answers immediately. The systems want the same outcome and that is to be efficient and product the information that it needs as quickly as possible. The systems are looking to be efficient, some systems can be looked at as more efficient than others and those are the systems that the companies will choose to house all their information and perform all the necessary tasks that they my have. The genetic algorithms seems as though it could potentially take the longest based on the time that it could have to wait to produce an answer while the neural network could take less time because neural systems are just a stepping stones into artificial networks. Expert and Agent-based systems seem to be the ones that are most commonly used by most organizations and business. They seem to house the most resources that are needed.
If I was selling baseballs, the agent-based technologies would be useful in deciding what direction to take with selling baseballs, mainly the data-mining agent. I believe that I could potentially have the same amount of information that Wal-Mart has on it's customers. Basic things such as name, phone number, address are a given. Information regarding purchases as well as transaction information would be needed in order to properly evaluate the finances of the company. I wouldn't immediately have as much information but with time I could potentially have it. This is true because we are both selling a product and the same information is needed in order to do so.
Faith Brown
References

"AI: Genetic Algorithms." AI: Genetic Algorithms. N.p., n.d. Web. 09 Sept. 2014. <http://biology.kenyon.edu/slonc/bio3/AI/GEN_ALGO/gen_algo.html>.
Engelmore, Robert S. "EXPERT SYSTEMS AND ARTIFICIAL INTELLIGENCE." EXPERT SYSTEMS AND ARTIFICIAL INTELLIGENCE. WTEC Hyper-Librarian, 1993. Web.
06 Sept. 2014. <http://www.wtec.org/loyola/kb/c1_s1.htm>.
"Genetic Algorithms." Introduction to. N.p., n.d. Web. 09 Sept. 2014. <http://www.doc.ic.ac.uk/~nd/surprise_96/journal/vol1/hmw/article1.html>.
Haag, Stephen, and Maeve Cummings. Management Information Systems for the Information Age. New York, NY: McGraw-Hill Irwin, 2013. Print.
"Types of Artificial Intelligence Systems." EHow. Demand Media, 14 Apr. 2011. Web. 08 Sept. 2014. <http://www.ehow.com/info_8222665_types-artificial-intelligence-systems.html>.
"What Is the Difference between Artificial Intelligence and Neural Networks?" Techopedia.com. N.p., n.d. Web. 09 Sept. 2014. <http://www.techopedia.com/2/27888/programming/what-is-the-difference-between-artificial-intelligence-and-neural-networks>.