The Value of Relevance

Robert Korsan

The highest cost technical activity in decision analysis, knowledge engineering and expert systems development is probabilistic assessment. Influence diagram (ID) technology has reduced assessment and improved decision models dramatically. However, once an ID has been assessed from the decision-maker (DM) and deterministic sensitivity analysis performed, a method to price the relevance arrows in the ID would allow the DM to balance the cost of assessment against the improvement in the authenticity of the value lottery and accuracy of the recommended action. This paper outlines such a method to price relevance arrows in an ID contianing discrete uncertainties and discrete decisions. The method also determines the sensitivity of a decision to the probability of particular outcomes of any uncertain event.


This page is copyrighted by AAAI. All rights reserved. Your use of this site constitutes acceptance of all of AAAI's terms and conditions and privacy policy.