|
| |
expert systems |
|
| |
|
|
| |
Software packages that attempt to mimic the behaviour of an expert in a given field. The notion that a computer could emulate an expert is attractive, since it implies that such expert systems could be used to standardize many decision-making processes, reduce errors, and improve the performance of regulatory agencies. An expert system consists of a rule base, containing a digital representation of the known decision rules, and a processor to evaluate the rules in a given instance and reach an appropriate decision. The builder of an expert system will commonly make use of a set of generic tools known as a \'shell\', which takes care of the storage of the rule base, and the processing functions. Expert systems have often been coupled with geographic information systems (GIS), which provide many of the inputs and may also display and manage the outputs, in supporting decision-making in forestry, resource management, hydrology, and urban and regional planning. The success of expert system approaches depends on the degree to which all relevant rules can be expressed in the highly constrained forms required of a rule base, and on the credibility of an expert system\'s output to its users. Rules are often imprecise or \'fuzzy\', and expert systems based on fuzzy reasoning have become popular in many geographic applications. Finally, an expert system can provide a useful formal structure for studying decision-making processes. (MG)
Suggested Reading Kim, T.J., Wiggins, L.L. and Wright, J.R., eds, 1990: Expert systems: applications to urban planning. New York: Springer-Verlag. |
|
| |
|
|
| |
Bookmark this page:
|
|
| |
|
|
| |
<< former term |
|
next term >> |
|
|
|
|
|
| |
|
|
|
