Expert Systems

1100 Words5 Pages
Being an “expert” means having a special knowledge or ability that allows you to perform actions skillfully, and at a higher level of confidence. It means having an extensive and task-specific skill learned from training, doing, and flat out experience. Knowing the facts, theories, and procedures about a general problem area is defined as being skilled in that area. A new subset of artificial intelligence was created with the definition of expert in mind. Early work with artificial intelligence was focusing too much on methods of search and heuristics which limited the search for solving problems. With much success, in 1978 some remarkable people at Stanford University designed a system called DENDRAL. It uses a chemist’s knowledge to find out what a chemical substance is. This was the birth of what we now call an “Expert System”. An expert system is defined as “a program that attempts to mimic human expertise by applying inference methods to a specific body of knowledge” (Darlington). “It is also said to be defined as an interactive computer based tool that will use both heuristics and facts to solve hard problems, based on the knowledge received from an expert”( The power of computers has exponentially grown to the point of infinite possibilities. This allows us to utilize them for increasing demanding tasks that can make life easier. An expert system does just that; it allows for computers to make automated decisions without human intervention or labor input. The field of artificial intelligence is right where these systems stem from, but differs in many ways from other artificial intelligence based methods like genetic algorithms. For one, like genetic algorithms, traditional computer software will use an alg... ... middle of paper ... ... system as a declarative format so the expert system can understand and interpret it. These people are inputting into the system while there are others who will use the system, which are called users. Knowledge engineers don’t have to just have the tech knowledge to code the system; they need to also have a creative ability. Being creative allows them to not just focus on the raw knowledge of the subject, but rather create an intelligent structure of that knowledge. The knowledge engineer most often works directly with the domain expert when creating a system. The reasons behind this relationship are simple; the domain expert provides the knowledge that is needed to be represented in the system. Structuring the reasoning correctly in the system is a must, which is why the engineer closely watches the domain experts’ decisions as they solve a question on their own.
Open Document