Copyright © 1992 Published by Elsevier Science Inc.
A computational normative theory of scientific evidence
Received 1 July 1990;
Abstract
A scientific reasoning system makes decisions using objective evidence in the form of independent experimental trials, propositional axioms, and constraints on the probabilities of events. I propose a collection of algorithms that derive probability intervals and estimate conditional probabilities from objective evidence in those forms. This reasoning system can manage uncertainty about data and rules in a rule-based expert system. I expect that the system will be particularly applicable to diagnosis and analysis in domains with a wealth of experimental evidence such as medicine. The algorithms currently apply to systems with arbitrary amounts of experimental evidence but with less than 20 variables. I discuss limitations of this solution and propose future directions for this research. This work can be considered a generalization of Nilsson's “probabilistic logic” to intervals and experimental observations.
Author Keywords: interval probability; evidence combination; experimental evidence; probabilistic logic; statistical inference






E-mail Article
Add to my Quick Links

Cited By in Scopus (0)






