|
In this article:
- What is an expert system
- Advantages and disadvantages of an expert system
- Uses of expert systems
What is an expert system?
An expert system or knowledge based system is a computer programming or collection thereof that has information regarding a particular subject. Expert systems are an avenue of artificial intelligence and have been around since the 1960s. Commercial applications of such systems began in the 1980s. Expert systems try to draw conclusions or inferences by analyzing information which is provided by the user. This information pertains to a particular subject alone that can be solved by a corresponding expert system. Thus, expert systems essentially work on the basis of logical reasoning. Expert systems may also be implemented in the form of a wizard whereby they provide the solution to a problem with the help of predefined steps, collecting information from the user at each step.
Advantages and disadvantages of an expert system
As explained above, expert systems draw conclusions based on logical reasoning. The use of expert systems has its own advantages and disadvantages.
Some of the important advantages of an expert system are:
- Expert systems can answer routine and repetitive questions in a short amount of time
- Expert systems can store high amounts of important subject-specific information
- Expert systems can help make the decision making process centralized
- Expert systems help decrease errors in a particular task which may arise due to human inefficiency
- Expert systems can detect the smallest details which may be undetectable by humans
However, expert systems have some limitations or disadvantages:
- Expert systems cannot apply common sense which may be necessary to take some decisions
- Expert systems cannot react creatively to uncertain situations like humans
- Expert systems are complex and difficult to program
- Expert systems are not versatile and cannot adapt themselves to changing environments all by themselves
Uses of Expert System
Expert systems find a variety of uses in the field of medicine, accounting, human resource management and production management. The development of an expert system for a particular purpose involves a long process of drafting, designing and programming the system.
Expert systems are probably the most successful variation of Artificial Intelligence but still, their use is very limited. Expert systems are known to commit errors that can be easily detected by a human. In the end, expert systems are nothing but advanced use of decision logic that has been longed used by programmers.
A typical expert system that has been used by people all over the world for a long time is the Microsoft Windows troubleshooting software which is part of the help system of the operating system. This software helps the ordinary user who does not posses advanced information about the working of the operating system to solve problems that may arise when using it by asking a series of questions, detecting the problem and explaining the required action to be taken by the user. This software provides information in the form of advice, suggestions and solutions.
Expert systems have also been used by computer games in the 1970s and 1980s as a variant of early Artificial Intelligence systems. Since the introduction of computer games, expert systems that are part of the computer game itself have been used to suggest logical moves or strategies when the player is playing against an artificial computer opponent. These moves, such as those in the game of chess, are result of logical deductions drawn out by analyzing the current condition of the game. This has been put to use in the baseball games such as Earl Weaver Baseball. Here, the computer would try to counter the moves made by the player based on analysis and logical reasoning. However, these types of uses are nothing but ‘primitive’ and today are termed as being simple AI. A popular use of expert systems today is for the detection of diseases in the field of medical science by asking the patient a set of questions and analyzing the symptoms in the answers given by him or her.
|
