How to reason with uncertain knowledge

Research output: Chapter in Book/Report/Conference proceedingChapterAcademic

Abstract

In this paper a reasoning process is viewed as a process of constructing a partial model of the world we are reasoning about. This partial model is a syntactic representation of an epistemic partial semantic model. In such a partial model different views on the world we are reasoning about can be represented. Multiple views can be the result of updating a partial model with information containing a disjunction or information described by sentences like: ‘most humans have brown eyes’. If a proposition does not have to hold in every view, we cannot be certain of it. To express this two certainty measures for conclusions based on a partial model will be defined, a probability and a likelihood measure. The former will be used for conclusions that express an expectation while the latter will be used for conclusions that express an explanation. To show the value of the approach various applications of these certainty measures will be described.key-wordmanagement of uncertainty.
Original languageEnglish
Title of host publicationUncertainty in Knowledge Bases, LNCS 521
PublisherSpringer Verlag
Pages403-412
Number of pages10
DOIs
Publication statusPublished - 1991

Cite this