What is an expert system? The most recent addition to the circle of information systems is the expert system. Expert systems are associated with an area of research known as artificial intelligence. We introduced expert systems and artificial intelligence in the World of computers. ” Artificial intelligence is the ability of a computer to reason, to learn, to strive for self-improvement, and to simulate human sensory capabilities. Like the DSS, expert systems are computer-based systems that help managers resolve problems or make better decisions.
However, an expert system does this with a decidedly different twist. It is an interactive computer-based system that responds to questions, asks for clarification, makes recommendations, and generally helps the user in the decision-making process. In effect, working with an expert system is much like working directly with a human expert to solve a problem because the system mirrors the human thought process. It even uses information supplied by a real expert in a particular field such as medicine, taxes, or geology.
An expert system applies preset IF-THN rules to solve a particular problem, such as determining a patient’s illness. Like management information systems and decision support systems, expert systems rely on factual knowledge, but expert systems also rely on heuristic knowledge such as intuition, judgment, and inferences. Both the factual knowledge and the heuristic rules of thumb are acquired from a domain expert, an expert in a particular field, such as jet engine repair, life insurance, or property assessment.
The expert system uses this human-supplied knowledge to model the human taught process within a particular area of experience .
Once completed, a knowledge-based system can approximate the logic of a well-informed human decision maker. Technically speaking; an expert system is the highest form of a knowledge-based system. In practice, the terms expert system and knowledge-based systems are used interchangeably. The less sophisticated knowledge based system is called assistant systems. An assistant system helps users make relatively straightforward decisions.
Assistant systems are usually implemented to reduce the possibility that the end user will make an error in judgment rather than to resolve a particular problem. In effect, expert system simulates the human thought process. To varying degrees, they can reason, draw inferences, and make judgments. Here is how an expert system works. Let’s use a medical diagnosis expert system as an example. Upon examining a patient, a physician might use an expert diagnosis system to get help in diagnosing the patient’s illness or, perhaps, to get a second opinion.
First the doctor would relate the symptoms to the expert system: male, age 10, temperature of 103, and swollen glands about the neck. Needing more information, the expert system might ask the doctor to examine the parotid gland for swelling. Upon receiving an affirmative answer, the system might ask a few more questions and even ask for lab reports before giving a diagnosis. A final question for the physician might be whether the patient had been previously afflicted with or immunized for parotitis. Based on the information, the expert system would diagnose the illness as parotitis, otherwise known as the mumps.
In recent years expert systems have been developed to support decision makers in a broad range of disciplines, including medical diagnosis, oil exploration, financial planning, tax preparation, chemical analysis, surgery, locomotive repair, weather prediction, computer repair, troubleshooting satellite, computer systems configuration, nuclear-power plant operation, newspaper layout, interpreting government regulations, and many others.
The benefits of an expert system are somewhat different from those of other decision support system s and of management information systems. An expert system enables the knowledge of experts to be canned, so to speak, the specialized knowledge of human experts can be captured in the form of an expert system. For example, at camp-bell’s soup company, Aldo cimino was the only expert trouble-shooter for Campbell’s giant cookers. He and his 43 years of experience were about to retire, so Campbell’s executives decided to drain his brain into an expert system. Mr. Cimino may be retired, but Campbell’s soup company continues to benefit from his years of experience. A single expert system can expand the decision-making capabilities of many people. In effect, an expert’s knowledge can be distributed to and used by anyone associated with a specific decision environment. For example, a number of loan officers at a bank can enlist the aid of an expert system for guidance in approving and rejecting loan applications. ?An expert system can improve the productivity and performance of decision makers. By having ready access to an electronic partner with vast expertise in a particular area, decision makers can progress more rapidly to the most acceptable solution. An n expert system can provide stability and consistency to a particular area of decision making. Unlike human beings, an expert system is consistent with regard to decision making. That is an expert system will always render the same decision for a given set of information. When people in similar decision-making situations (for example, commercial loans officers in a bank) have access to the advice and guidance of an expert system, the decisions they make tend to be consistent with one another. ?An expert system reduces dependencies on critical personnel.
Human beings retire, get sick, take vacations, and only a few of them ever attain the status of expert. Computers do not take coffee breaks. Expert system can benefit from the experience, immediately and after they retire. ?An expert system is an excellent training tool. Companies are using expert systems to train decision makers in a way similar to airlines’ use of flight simulators to train pilots. During training, individuals work through a particular decisions with an expert system. After making the decision, they review the documentation of the decision rationale generated by the expert system.
From this documentation, they learn how decisions are made within the context of a particular environment.
When we talk about expert system, we usually mean systems that can help decisions makers working in a particular domain of expertise, such as the configuration of computer system or commercial lending. As mentioned earlier, these expert systems are the result of substantial development efforts. The software that enables the development of this expert has no “intelligence” and is known as the expert system shell.
Expert system shells are usually domain-independent proprietary software packages that have no application “knowledge”. An expert system shell contains the generic parts needed to create an expert system for a specific application. For example, the expert system provides companies with the capabilities needed to construct a knowledge base and the facility by which the user interacts with knowledge base. The primary components of the expert system shell are the knowledge-acquisition facility, the knowledge base, the inference engine, and the user interface as shown in fig. . Knowledge Domain Engineer expertRules of thumb Facts MIS AND DSS APP. fig. 2: Components of an Expert System Shell ?Knowledge-acquisition facility: the knowledge- acquisition facility is that component of the expert system shell that permits the construction of the knowledge base. The knowledge base is created through the cooperative efforts of a knowledge engineer and one or more experts in a particular field called domain expert: The knowledge engineer translates the expert’s knowledge into factual knowledge and rules to crate the knowledge base. Knowledge base: Appropriate facts and rules are entered into the knowledge base during acquisition phase. To complete the knowledge base, the knowledge engineer, in cooperation with the domain expert, enters the following: the identification of the problems to be solved; possible solution to the problems; and how to progress from problem to solution (primarily through facts and rules of inference). Facts (employee name and so on) needed to articulate the solutions to the user are retrieved from the corporate database. Inference engine: The inference engine is the nucleus of an operational expert system. It is the vehicle by which the facts and rules in the knowledge base is applied to a problem. The inference engine gives an expert system its ability to reason. It does this by leading the user through a logic path that result in a solution. ?User interface: Heuristic procedures are informal; that is, there are no formal algorithms available to solve the problem. An expert system problem is addressed by one strategy as long as it looks promising.
The system always retains the option to switch to another strategy. This heuristic approach requires a flexible user interface. This component of an expert system enables the type of interaction between end user and expert system needed for heuristic processing. The user interface permits the end user to describe the problem or goal. It permits both the end user and the expert system to structure questions and responses. Along with a response to a particular inquiry, an expert system usually explains and documents the rationale of why a particular course of action was recommended. .
One of the myths surrounding expert system is that they will replace human experts. While expert systems augment the capabilities of humans and make them more productive, they will never replace them. Expert system and human complement one another in the decision-making process. The computer –base expert system can handle routine situations with great accuracy, thereby relieving someone of the burden of a detailed manual analysis. Humans can combine the insight of an expert system with their flexible intuitive abilities to resolve complex problems.
The number and variety of expert system applications have increased dramatically with the advent of powerful, cost-effective microcomputers. Expert system advise financial analysts on the best mix of investment; help taxpayers interpret the tax laws; help computer repair persons diagnose the problem of malfunctioning computer; and help independent insurance agents select the best overall coverage for their business clients. In the short period of their existence, expert systems have operated impressively and they continue to improve. Decisions making in every environment are developing or contemplating developing an expert system.
Attorneys will hold mock trials with expert system to “pre-try” their cases Doctors routinely will asked a second opinion. Architects will “discuss” the structural designed of a building with an expert system. Military officers will “talk” with the “expert” to plan battlefield strategy. City planner will “ask” an expert system to suggest optimal locations for recreational facilities. Some computer industry observers believe that expert systems are the wave of the future and that each of us will have “expert” help and guidance in our respective professions.