Belief Revision in a Deductively Open Belief Space

Abstract

I am researching the traditional belief revision integrity constraints and postulates, which are designed for deductively closed belief spaces, and revising them so that they are applicable to implemented knowledge representation and reasoning systems with deductively open belief spaces (DOBS). A knowledge representation and reasoning system must be able to deal with contradictions and revise beliefs. This is especially important to data fusion, where information is combined from multiple sources, which might contradict each other. Most theoretical postulates for belief revision and belief contraction assume a deductively closed belief space (DCBS), where all beliefs derivable from a belief space are in that belief space. This is hard (or impossible) to produce in an implemented belief revision system,