REPRESENTATION, PROPAGATION AND COMBINATION OF UNCERTAIN INFORMATION

This study examines the fundamental issues involved in the quantitative and qualitative representation, propagation and combination of uncertain information from different sources. Various classes of quantitative uncertainty measures and qualitative uncertainty judgments are reviewed and their compa...

Full description

Saved in:
Bibliographic Details
Published in:International journal of general systems 1994-10, Vol.23 (1), p.59-83
Main Authors: YAO, Y. Y., WONG, S. K. M.
Format: Article
Language:English
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This study examines the fundamental issues involved in the quantitative and qualitative representation, propagation and combination of uncertain information from different sources. Various classes of quantitative uncertainty measures and qualitative uncertainty judgments are reviewed and their compatibilities are analyzed. Three different propagation paradigms are studied. They are the quantitative paradigm exemplified by the Bayes rule of conditionalization, the mixed paradigm derived from the compatibility view of belief functions, and the qualitative paradigm employed in the exposition of belief structures. The problem of consistency in these propagation paradigms is addressed. The propagation of uncertain information is discussed using the proposed paradigms. The process of pooling uncertain information is illustrated by the combination of belief functions and relations. A qualitative combination rule is proposed and is based on a well-defined distance function between belief relations. The results of this study are useful for the design of an inference network capable of performing both quantitative and qualitative inferences.
ISSN:0308-1079
1563-5104
DOI:10.1080/03081079408908030