Abstract
Deterministic rule-based expert systems, introduced in Chapter 2, do not deal with uncertainties because objects and rules are treated deterministically. In most practical applications, however, uncertainty is the rule not the exception. For example, a question that often arises in medical diagnosis is: Given that a patient has a set of symptoms, which disease is the patient most likely to have? This situation involves some degree of uncertainty because
-
The facts (concrete knowledge) may not be accurate. For example, a patient may not be sure whether or not he had a fever last night. Thus, there is a degree of uncertainty in the information associated with each patient (subjectivity, imprecision, lack of information, errors, missing data, etc.).
-
The abstract knowledge is not deterministic. For example, the relationships among diseases and symptoms are not deterministic because the same group of symptoms may be associated with different diseases. In fact, it is not uncommon finding two patients with the same symptoms but different diseases.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 1997 Springer-Verlag New York, Inc
About this chapter
Cite this chapter
Castillo, E., Gutiérrez, J.M., Hadi, A.S. (1997). Probabilistic Expert Systems. In: Expert Systems and Probabilistic Network Models. Monographs in Computer Science. Springer, New York, NY. https://doi.org/10.1007/978-1-4612-2270-5_3
Download citation
DOI: https://doi.org/10.1007/978-1-4612-2270-5_3
Publisher Name: Springer, New York, NY
Print ISBN: 978-0-387-94858-4
Online ISBN: 978-1-4612-2270-5
eBook Packages: Springer Book Archive