24.11.2014 Views

flex Expert System Toolkit - LPIS

flex Expert System Toolkit - LPIS

flex Expert System Toolkit - LPIS

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

Appendix D - Dealing with Uncertainty 226<br />

Appendix D - Dealing with Uncertainty<br />

This document aims to serve as a basic introduction to the various<br />

techniques that Flex now offers to support uncertainty. For a more detailed<br />

explanation, you are referred to:<br />

“Knowledge-Based <strong>System</strong>s for Engineers and Scientists“<br />

( Hopgood, CRC Press, ISBN: 0-8493-8616-0 )<br />

Traditional expert systems work on the basis that everything is either true or<br />

false, and that any rule whose conditions are satisfiable is useable, i.e. its<br />

conclusion(s) are true. This is rather simplistic and can lead to quite brittle<br />

expert systems. Flex now offers support for where the domain knowledge is<br />

not so clearcut.<br />

Given a rule:<br />

rule1: if A & B then C<br />

there are 3 potential areas for uncertainty.<br />

- Uncertainty in data (how true are A and B)<br />

- Uncertainty in the rule (how often does A and B imply C)<br />

- Impreciseness in general<br />

The first 2 can be handled using probabilities and the third using fuzzy logic.<br />

Uncertainty in Data<br />

Combining Probabilities<br />

A probabilistic rule in <strong>flex</strong> can look like a production rule:<br />

uncertainty_rule r33<br />

if the temperature is high<br />

and the water_level is not low<br />

then the pressure is high .<br />

Later on we shall see that this is equivalent to:<br />

6<br />

<strong>flex</strong> toolkit

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!