Contextual Argumentation in Ambient Intelligence

摘要

The imperfect nature of context in Ambient Intelligence environments and the special characteristics of the entities that possess and share the available context information render contextual reasoning a very challenging task. Most current Ambient Intelligence systems have not successfully addressed these challenges, as they rely on simplifying assumptions, such as perfect knowledge of context, centralized context, and unbounded computational and communicating capabilities. This paper presents a knowledge representation model based on the Multi-Context Systems paradigm, which represents ambient agents as autonomous logic-based entities that exchange context information through mappings, and uses preference information to express their confidence in the imported knowledge. On top of this model, we have developed an argumentation framework that exploits context and preference information to resolve conflicts caused by the interaction of ambient agents through mappings, and a distributed algorithm for query evaluation.