Handling Incoming Beliefs

摘要

Most logic-based approaches to knowledge and belief change in artificial intelligence assume that when a new piece of information comes up, it should be merely added to the current beliefs or knowledge when this does not lead to inconsistency. This paper addresses situations where this assumption does not hold. The focus is on the construction of Boolean standard-logic knowledge and belief bases in this context. We propose an approach to handle incoming beliefs that can require some formulas reconstruction or a form of preemption to be performed.