What is an expert system?

Answered by Priya Chetty on January 7, 2012 Leave a comment (0) Go to comments

An expert system depends on a knowledge codification obtained from human experts to reproduce human expertise in a limited domain and thus offer an expert views to queries in the domain that might be imposed by a non-expert. An expert system does not arrive at an expert views using similar processes as a human does. This is because present expert systems are not unable to reason from first principles, common sense or analogies nor can an expert system perceive qualitative significance and jump to conclusions intuitively (Patridge and Hussain, 1992, p 149-150). The below figure shows the expert systems:

Figure 1: Expert Systems. Source: Waterman (1986), A Guide to Expert Systems, Addisson Wesley, Melbourne, p 18

Stevens (1984, p 40) gives the following definition that expert systems are machines that reason and think as an expert would in a specific domain. For instance a medical diagnoses expert system may request as input the symptoms of patients, other similar facts and test results using these as pointers it may search its database for information that may lead to the recognition of illness. A true expert system not only performs the traditional computer functions of managing huge number of data but it also manipulates that data so the result is a meaningful answer to a less than wholly specified query. An expert system can be also defined as a computer system that simulates human experts in a given area of specialization (Stevens (1984, p 40) cited in Castillo, Gutierrez and Hadi (1997, p 2).

Reference

Patridge D and Hussain K M (1992), Artificial intelligence and business management, Ablex Publishing, USA, p 149-150.
Castillo E, Gutierrez J M and Hadi A S (1997), Expert systems and probabilistic network models, Springer, New York, p 2.