Knowledge Acquisition for Probabilistic Expert Systems

Reference: Lehmann, H. P. Knowledge Acquisition for Probabilistic Expert Systems. 1988.

Abstract: Recent interest in probability-based expert systems has focused on the potential these systems have for being coherent with the beliefs of the modeled expert or of the user and consistent given any set of evidence. We have used the probabilistic formalism in creating the REFEREE system, a belief-network-based expert system designed to aid readers in determining the credibility of a randomized clinical trial. In this paper, we explore the effect the formalism had on the process of knowledge acqisition based on this experience. Although the system is still in development, we can report several of those effects. Specifically, the need to make operational definitions of concepts deemed important to the expert forced us to organize a domain that was formulated initially for a rule-based system. Categorizing probability distributions as being logical, probabilistic, or prototypical helped us to decrease the number of probability assesments. On the other hand, the lack of an intermediate prototype may have prolonged development, and computational limitations forced occasional compromises. The reality of building expert systems in a probabilistic paradigm may not be as hard as some critics have predicted.

Jump to... [KSL] [SMI] [Reports by Author] [Reports by KSL Number] [Reports by Year]
Send mail to: ksl-info@ksl.stanford.edu to send a message to the maintainer of the KSL Reports.